Evaluating Machine Learning Models on the Ethereum Blockchain for Android Malware Detection

Android, the most popular mobile operating system, with billions of active users and more than 2 million apps, has motivated advertisers, hackers, fraudsters and cyber-criminals to develop malware of all types for it. In recent years, extensive research has been conducted on malware analysis and detection for Android devices, even though Android has already implemented various security mechanisms to deal with the problem. In this paper, we developed a consortium blockchain network to evaluate various machine learning models for a given malware dataset. A reward is offered using smart contracts as an incentive to the competitors for their work by allowing them to submit solutions through training with selected machine learning models in a secure and trustworthy manner. The analysis of datasets by competitors helps various organizations in the network to enhance or boost their current malware detection or defense tools. The decentralized network provides transparency, enhances security and reduces the cost in managing all relevant data by eliminating third parties. We used DREBIN dataset in the developed framework for initial experiments and the encouraging results are presented.

[1]  Pierre Geurts,et al.  Extremely randomized trees , 2006, Machine Learning.

[2]  Byung-Gon Chun,et al.  TaintDroid: An Information-Flow Tracking System for Realtime Privacy Monitoring on Smartphones , 2010, OSDI.

[3]  Mohammed Ouzzif,et al.  A Novel Security Framework for Managing Android Permissions Using Blockchain Technology , 2018, Int. J. Cloud Appl. Comput..

[4]  Neil Wilson,et al.  Decentralised firewall for malware detection , 2017, 2017 International Conference on Advances in Computing, Communication and Control (ICAC3).

[5]  Andrew H. Sung,et al.  Evaluation of Tree Based Machine Learning Classifiers for Android Malware Detection , 2018, ICCCI.

[6]  Andrew H. Sung,et al.  Malware Analysis on Android Using Supervised Machine Learning Techniques , 2018 .

[7]  Yajin Zhou,et al.  RiskRanker: scalable and accurate zero-day android malware detection , 2012, MobiSys '12.

[8]  R. K. Agrawal,et al.  Relevant Feature Selection from a Combination of Spectral-Temporal and Spatial Features for Classification of Motor Imagery EEG , 2018, Journal of Medical Systems.

[9]  Konrad Rieck,et al.  DREBIN: Effective and Explainable Detection of Android Malware in Your Pocket , 2014, NDSS.

[10]  Yajin Zhou,et al.  Hey, You, Get Off of My Market: Detecting Malicious Apps in Official and Alternative Android Markets , 2012, NDSS.

[11]  Steve Hanna,et al.  Android permissions demystified , 2011, CCS '11.

[12]  Eric Filiol,et al.  Developing a Κ-ary malware using blockchain , 2018, NOMS 2018 - 2018 IEEE/IFIP Network Operations and Management Symposium.

[13]  Charles Noyes.  BitAV: Fast Anti-Malware by Distributed Blockchain Consensus and Feedforward Scanning , 2016, ArXiv.

[14]  Heng Yin,et al.  DroidScope: Seamlessly Reconstructing the OS and Dalvik Semantic Views for Dynamic Android Malware Analysis , 2012, USENIX Security Symposium.

[15]  Arun Kumar Sangaiah,et al.  Root Exploit Detection and Features Optimization: Mobile Device and Blockchain Based Medical Data Management , 2018, Journal of Medical Systems.

[16]  A. Besir Kurtulmus,et al.  Trustless Machine Learning Contracts; Evaluating and Exchanging Machine Learning Models on the Ethereum Blockchain , 2018, ArXiv.

[17]  Andrew H. Sung,et al.  Evaluating Machine Learning Models for Android Malware Detection: A Comparison Study , 2018, ICNCC.

[18]  Xiaojiang Du,et al.  Consortium Blockchain-Based Malware Detection in Mobile Devices , 2018, IEEE Access.

[19]  Patrick D. McDaniel,et al.  On lightweight mobile phone application certification , 2009, CCS.