ABSTRACT:With the development of internet era Smartphone’s have become more popular allaround the world. Android is the most popular mobile operating system. With theincreasing use of Smartphone’s the number of malwares attacking theseSmartphone’s have also been increased. To effectively detect the new malwaresand malicious software variants has been a difficult problem. Our method usesthe Keywords Correlation Distance to compute the correlation between key codessuch as API calls, Android permissions, the common parameters, and the commonkeywords in Android malware source code. Then Support Vector Machine is appliedto make the system gain to accommodate the function of the new malicioussoftware sample, so as to detect new malicious software and existing malwares.This method is different from the conventional methods which are based on thecontext of the text. This method combines the Characteristics Of The Malicious SoftwareCategories And Operating Environment To Record The Behaviour Of The MaliciousSoftware.
Keywords:Androidmalware, Machine learning, Keywords Correlation Distance, SVM