PAIRED: An Explainable Lightweight Android Malware Detection System

Mohammed M. Alani, Ali Ismail Awad

Research output: Contribution to journalArticlepeer-review

25 Citations (Scopus)


With approximately 2 billion active devices, the Android operating system tops all other operating systems in terms of the number of devices using it. Android has gained wide popularity not only as a smartphone operating system, but also as an operating system for vehicles, tablets, smart appliances, and Internet of Things devices. Consequently, security challenges have arisen with the rapid adoption of the Android operating system. Thousands of malicious applications have been created and are being downloaded by unsuspecting users. This paper presents a lightweight Android malware detection system based on explainable machine learning. The proposed system uses the features extracted from applications to identify malicious and benign malware. The proposed system is tested, showing an accuracy exceeding 98% while maintaining its small footprint on the device. In addition, the classifier model is explained using Shapley Additive Explanation (SHAP) values.

Original languageEnglish
Pages (from-to)73214-73228
Number of pages15
JournalIEEE Access
Publication statusPublished - 2022


  • Android
  • machine learning
  • malware
  • malware detection
  • XAI

ASJC Scopus subject areas

  • General Computer Science
  • General Materials Science
  • General Engineering
  • Electrical and Electronic Engineering


Dive into the research topics of 'PAIRED: An Explainable Lightweight Android Malware Detection System'. Together they form a unique fingerprint.

Cite this