ANDROID Malware Detection And Implementation Based On Reverse Engineering

The number of android malware application is increasing in the new future.Smartphone,tablets and other plays an increasingly important role in people's life.Android has become the largest installed smartphone operating system.In all aspects of people's lives are filled with a large number of different types of different functions with Android applications.Bring convenience to people's basic necessities of life.At the same time,all kinds of applications makes the user difficult to distinguish the authenticity.Led to the Android smartphone operating system security situation become more and more severe.Strengthen the detection of Android applications,to prevent criminals steal the user information,It is imminent to protect the Android system security.Experts at home and abroad have done a lot of research on the security of Android system.Because the static detection of Android applications do not need to implement the Android program,At the beginning of analysis the android applications users can download applications or install applications.Compared with the dynamic detection The static detection of android applications has less time-consuming and less systematic resources,higher efficiency.The application of the unknown malicious intrusion has certain ability of prevention,people pay more attention to it.This thesis based on Android application software development principles,achieve the Android malware detection system.This system implements the reverse analysis of Android application,get the authority information of Android application and the application programming interface(API).Combining Permission Information and Sensitive API Information to Establish Malware Feature Information Database.And improve the naive Bayesian classification algorithm in machine learning,Different weights are assigned to different privilege information attributes.The system adds the accuracy of the detection by using a classification algorithm which based on attribute weighting and combining the permission information with the API call.