A Survey of Android Malware Detection Strategy and Techniques

The expeditious growth of Android malwares has posed a serious challenge in front of researchers. The researchers are continuously proposing countermeasures and developing tools to mitigate against such attacks. In this paper, widely used techniques that have been proposed recently by researchers have been explored. The key contributions of each of these techniques along with their limitations have been analyzed. All these techniques were compared based on nine parameters and it was identified that Mobile Sandbox tool is the best when time factor is not considered because it possess the capability of both a static and dynamic analysis, native API call tracking and web accessibility. If time factor is considered, then Dendroid performs best among all. This is due to the reason that it applies text mining to get the signature of malware and it can also classify unknown malware sample through 1-NN classifier.