Steven Gianvecchio, Detecting Bots in Online Games using Human Observational Proofs

The abuse of online games by automated programs, known as bots, hasgrown significantly in recent years. The conventional methods fordistinguishing bots from humans, such as CAPTCHAs, are not effective ina gaming context. This talk presents a non-interactive approach based onhuman observational proofs for continuous game bot detection. HOPsdifferentiate bots from human players by passively monitoring inputactions that are difficult for current bots to perform in a human-likemanner. The talk describes a prototype HOP-based game bot defense systemthat analyzes user-input actions with a cascade-correlation neuralnetwork to distinguish bots from humans. The experimental results showthat the HOP system is effective in capturing game bots in World ofWarcraft, raising the bar against game exploits and forcing attackers tobuild more complicated bots for detection evasion in the future. About the speaker: Steven Gianvecchio received his Ph.D. in Computer Science from theCollege of William and Mary in 2010. He is a Senior Scientist at theMITRE Corporation, McLean, VA. His research interests include networks,distributed systems, network monitoring, intrusion detection, traffic modeling, and covert channels.