On Detecting Cryptojacking on Websites: Revisiting the Use of Classifiers

Area of Science:

• Cybersecurity
• Machine Learning
• Blockchain Technology

Background:

• Cryptojacking involves malware secretly using a victim's computational resources for cryptocurrency mining.
• The profitability of cryptocurrencies has led to a rise in cryptojacking attacks, which are often undetected by users.
• Detecting and blocking cryptojacking is a growing research area within cryptocurrency and blockchain technology.

Purpose of the Study:

• To explore and evaluate multiple Machine Learning (ML) classification models for detecting cryptojacking on websites.
• To identify the most effective features for predicting cryptojacking using feature selection methods.
• To compare the performance of various ML models against each other and against existing Deep Learning approaches.

Main Methods:

• Utilized a dataset comprising network and host features for cryptojacking detection.
• Applied feature selection techniques, including statistical methods (e.g., Test Anova) and wrapper methods, to reduce model complexity and identify predictive features.
• Trained and evaluated several ML classification models: Logistic Regression, Decision Tree, Random Forest, Gradient Boosting Classifier, k-Nearest Neighbor, and XGBoost.

Main Results:

• Simple ML models, including Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, and k-Nearest Neighbor, demonstrated high success rates in detecting cryptojacking.
• These simpler models achieved performance comparable to or better than advanced algorithms like XGBoost.
• The findings indicate that basic ML models can be as effective, or even more so, than complex Deep Learning methods in cryptojacking detection.

Conclusions:

• Machine learning offers effective solutions for identifying cryptojacking threats.
• Simpler machine learning models provide a viable and efficient alternative to complex algorithms for cryptojacking detection.
• Further research can build upon these findings to enhance the security of web users against cryptojacking.