How to predict e-mail viruses under uncertainty
InSeon Yoo, Ulrich Ultes‐Nitsche · 2005
This paper answers (addresses) the questions on how to detect email viruses without signatures and how to determine the probability whether the mail is abnormal and how to detect virus patterns in an infected file. In order to find out relations between email viruses and detectable knowledge, we analysed propagation of email viruses and characteristics of email viruses, studied infected files' structures and applied Bayesian networks and self-organizing maps to adaptive detection against email viruses.