Machine learning for malware detection

• Main Author: Hong, Qing Fu
• Other Authors: Lin Shang-Wei
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Computer science and engineering
• Online Access: http://hdl.handle.net/10356/70351
• Institution: Nanyang Technological University

Smartphones have been an integral part of our daily lives today. From instant messaging to performing online banking, smartphones have brought tremendous convenience to the people but also an ever-increasing reliance on them. With Android smartphones having the largest user base in the smartphone market, Android applications have become a means for attackers to infect smartphones with malware in an attempt to gain benefits. Therefore, it is critical to be able to identify malware effectively. In this project, the focus will be to experiment the viability of system call graphs together with machine learning to construct a malware detector, aiming to classify if an android application is malicious or benign. Different machine learning algorithms will also be experimented and compared to evaluate their results. The report will conclude with recommendations for future work at the end.