Machine Learning Approaches to Detect Browser-Based Cryptomining

Xavier, Sherwin Norman (2020) Machine Learning Approaches to Detect Browser-Based Cryptomining. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (651kB) | Preview
Preview	PDF (Configuration manual) Download (549kB) | Preview

Abstract

Modern browsers provide support to multiple APIs that make use of JavaScript leading to an increase in developing useful web applications but also malicious activities. One such malicious activity is browser-based cryptomining. Browser-based cryptomining activities are a way to hijack a user’s system without any permission from the end-user. This is a result of a rise in the popularity of cryptocurrencies that support mining activities requiring a CPU. This study proposes two approaches, static analysis and dynamic analysis to detect browser-based cryptomining activities. The static analysis uses the complexity features of a JavaScript code to design, implement and evaluate three unsupervised machine learning models while the dynamic analysis uses the different performance parameters of a system to design, implement and evaluate three supervised machine learning models. Ultimately, One-Class SVM anomaly detection model performed well for the static analysis with an accuracy of 78.9% while KNN classification model performed well for the dynamic analysis with an accuracy of 98.8%. Matthew’s Correlation Coefficient statistical test results supported the results of this study. Keywords – Cryptojacking, Browser, Machine Learning, Static Analysis, Dynamic Analysis.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security
Divisions:	School of Computing > Master of Science in Cyber Security
Depositing User:	Dan English
Date Deposited:	27 Jan 2021 18:58
Last Modified:	27 Jan 2021 18:58
URI:	https://norma.ncirl.ie/id/eprint/4521

Actions (login required)

View Item