Home
Call For Papers
Submission
Author
Registration
Publications
About
Contact Us

  Machine Learning Based Android Malware Detection Using Permission and Function Call  
  Authors : Emi Johnson; Deepa Rajan S; Radhakrishnan B
  Cite as:

 

Over the last few years smart phones have brought peoples lives to a new high level. With this rapid development of smart phones, android a Linux based open source mobile operating systems popularity will also increase . Android is mainly designed for touch screen devices such as tablets, smart phones, etc. Also, it allows the use of third -party applications. Due to the widespread use of android operating system, it has become one of the favourite targets of cyber criminals. So an effective and efficient malware detection system is very important. This paper proposes a machine learning based technique that is capable of finding malware and benign applications. Here first we reduce the total number of permission of an android application using reduction techniques . This technique consist of three steps.1.Scoring permission using their negative values 2.Scoring permission using support 3.Permission mining using association rule. Using this reduction technique we find the most significant 22 permission . After reduction technique we analyse the function calls to permission controlled function. As a result, this technique provide a more efficient and effective result to determine whether an application is malware or benign.

 

Published In : IJCSN Journal Volume 8, Issue 3

Date of Publication : June 2019

Pages : 208-212

Figures :02

Tables : 01

 

Emi Johnson : is pursuing her M.Tech. degree on Computer Science and Engineering at APJ Abdul Kalam Technological University, Thiruvananthapuram. Her research interests are in Image processing, Data Mining and Data Security.

Deepa Rajan S : is working as Assistant Professor in Computer Science and Engineering Department. She has 10 years experience in teaching. Her research interests focuses on Data Security, Image Processing.

Radhakrishnan B : is working as the Head of CSE department. He has more than 14 years experience in teaching and has published papers on data mining and image processing. His research interests include image processing, data mining, image mining.

 

Android App, Android Permission, Malware, Benign

In this paper, we have shown that it is possible to reduce the number of permissions to be analysed for mobile malware detection, while maintaining high effectiveness and accuracy. To extract only signi?cant permissionsthroughasystematic,3-level of reduction technique is used. We have developed a malware detection system based on permission and API calls. Our approach performs as well as or better than techniques that consider more permissions or all permissions. By using significant permission, we can improve performance a lot.

 

[1] Merging Permission and API features for Android Malware detection ;Mengyu Qiao [2] Malware Detection Approach for Android systems Using System Call Logs; Sanya Chaba [3] DREBIN: Effective and Explainable Detection of Android Malware in Your Pocket; Daniel Arp [4] Rough Droid: Operative Scheme for Functional Android Malware Detection; KhaledRiad [5] W. Wang, X. Wang, D. Feng, J. Liu, Z. Han, and X.Zhang, "Exploring permissioninducedriskinandroidapplicationsformaliciousapplicatio n detection," Information Forensics and Security.