Auto-detection of sophisticated malware using lazy-binding control flow graph and deep learning

• Published in: Computers & security Vol. 76; pp. 128 - 155
• Main Authors: Nguyen, Minh Hai, Nguyen, Dung Le, Nguyen, Xuan Mao, Quan, Tho Thanh
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ltd 01.07.2018; Elsevier Sequoia S.A
• Subjects: Anti-virus software; Binary-based control; Computer programming; Computer viruses; Deep learning; Dynamically executed contents; Flow graph; Graphical representations; Image classification; Lazy-binding CFG; Learning; Malware; Metamorphic virus; Mutation; Packing techniques; Polymorphic virus; Program verification (computers); Software; Studies; Binary-based control; Deep learning; Packing techniques; Lazy-binding CFG; Polymorphic virus; Dynamically executed contents; Malware; Metamorphic virus; Mutation; Flow graph

To date, industrial antivirus tools are mostly using signature-based methods to detect malware occurrences. However, sophisticated malware, such as metamorphic or polymorphic virus, can effectively evade those tools by using some advanced obfuscation techniques, including mutation and the dynamicall...