Feature Extraction And Detection Of Malware Based On Similarity In Android Platform

With the advent of the Internet era,the popularity of smart phones in the world is also getting higher and higher,and Android smart phone operating system with its excellent performance,access to a huge market share.With the development of smart phones,more and more mobile phone malware also appeared in the market,harms to the information security of users.The major security laboratories also focus on security protection,but how to effectively killing the new malware and malicious software variants have been a problem.In view of the traditional feature extraction method based on binary program,this paper presents a method for feature extraction of JAVA source code.The method uses the Google distance computing the correlation between key codes such as API calls,Android permissions and the common parameters,and mining the common key words in Android malware source code.Then we can feature Android malware by classifying them according to their similarity and comparing the experimental results with the normal software.Then through SVM vector machine can make the system gain to accommodate the function of the new software virus sample,so as to achieve detecting new malicious software and old type of malware.This method breaks the conventional method,which is based on the context of the text,and combines the characteristics of the whole virus software operating environment to record the behavior of the virus.Experiments show that the method is effective.