Daftar Isi:
  • Malicious software development has become a serious threat to the security of android system, especially a new malicious software. There are new malicious software cannot be detected by conventional antivirus. This study created a framework for building software to detect and measure the accuracy of new and existing malware detection. In this study using combination of n-gram algorithm and backpropagation algorithm. N-gram algorithm is done statically so it needs a backpropagation algorithm so that the detection can be done dynamically. The application of combination n-gram algorithm and backpropagation algorithm successfully detect Droiddream type malware and this research can also detect malware variants, among other DroidKungfu, BaseBridge and DroidCoupon although it produce low accuracy. Therefore, this research can help in detecting a Droiddream, DroidKungFu, BaseBridge and DroidCoupon types of android malware on user devices for the safety and comfort of the user.