A new semantics-based android malware detection

With its high market share, the Android platform has become a growing target for mobile malware, which posed great threat to customers' safety. Meanwhile, malwares employed various techniques, take code obfuscation for example, to evade detection. The commercial mobile anti-malware products, however, are vulnerable to common code transformation techniques. This paper proposes an enhanced malware detection approach which combines advantage of static analysis and performance of ensemble learning to improve Android malware detection accuracy. The model extracts semantics-based features which can resist common obfuscation techniques, and also uses feature collection from code and app characteristics through static analysis. Real-world malware samples are used to evaluate the model and the results of experiments have proved that this approach improved the efficiency with AUC of 2.06% higher than previous approach.