JISE


  [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]


Journal of Information Science and Engineering, Vol. 36 No. 1, pp. 13-29


DroidSD: An Efficient Indexed Based Android Applications Similarity Detection Tool


JUNAID AKRAM, ZHENDONG SHI, MAJID MUMTAZ AND PING LUO
Key Laboratory of Information System Security
School of Software
Tsinghua University
Beijing, 100084 P. R. China
E-mail: fznd15; szd15; maji16g@mails.tsinghua.edu.cn; luop@mail.tsinghua.edu.cn


Android is becoming more and more popular in recent years. Meanwhile, it has been noticed that the security threats are also increasing with the passage of time because of the code reuse from other applications. In this paper, we propose and design DroidSD, a clone detection tool for Android applications, that helps to detect different types of code clones from APK's source code. A prototype has been developed and implemented to detect clones in Android applications. DroidSD detects Type-1, Type-2 and Type-3 (near-miss) clones in Android applications at the source code level with high accuracy rate, which was not possible in previous Android similarity detection techniques. DroidSD can detect full and partial level similarity between applications. We evaluate DroidSD on real time data-set and count the Recall and Precision on BigCloneBench, which is quite significant. The real time data-set includes 30,500 Android applications including top one, i.e. Chrome, Firefox, Gmail, WhatsApp, GoogleMap, Google-PlayStore, Baidu and BigCloneBench.


Keywords: clone detection, maintaining APK code, Android apps re-usability, plagiarism detection, Apps similarity detection, information security

  Retrieve PDF document (JISE_202001_02.pdf)