HIDE AND SEEK: AN ADVERSARIAL HIDING APPROACH AGAINST PHISHING DETECTION ON ETHEREUM
DOI:
https://doi.org/10.58885/ijcsc.v11i1.15.amKeywords:
Ethereum, blockchain security, phishing detection, machine learning, transaction analysis, Random Forest, Gradient Boosting, adversarial attacks.Abstract
The increasing use of Ethereum for cryptocurrency transactions and decentralized applications has created new security challenges, particularly phishing attacks involving fraudulent accounts and malicious transactions. Since blockchain transactions are generally irreversible, identifying suspicious account behavior at an early stage is important for reducing potential financial losses. This paper presents a machine learning-based framework for detecting phishing accounts using Ethereum transaction behavioral data. The framework performs data preprocessing, feature extraction, supervised classification, and performance evaluation using Logistic Regression, Support Vector Machine, Random Forest, and Gradient Boosting Decision Tree, using transaction-level attributes such as sender and receiver addresses, transferred value, and contract information. The models are evaluated using Accuracy, Precision, Recall, F1-Score, and confusion matrix analysis. Among the four classifiers, Random Forest achieves the strongest performance, with 97.19% accuracy and a 97.79% F1-score, while Logistic Regression, SVM, and Gradient Boosting Decision Tree provide comparatively lower but still meaningful detection performance. Feature-importance analysis is also used to identify the transaction characteristics that contribute most to phishing classification. In addition, adversarial hiding is considered to examine the robustness of the trained models when malicious behavior is deliberately modified to resemble legitimate activity; the resulting Attack Success Rate ranges from 68.14% for Logistic Regression to 100% for Random Forest and Gradient Boosting Decision Tree, showing that high accuracy under normal conditions does not guarantee resistance to adaptive attackers. The proposed framework provides a systematic approach for comparing machine learning models for Ethereum phishing detection and analyzing their robustness against evolving attack behavior, offering a foundation for developing more adaptive and reliable blockchain security systems.
References
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.
Chen, W., Guo, X., Chen, Z., Zheng, Z., & Lu, Y. (2020). Phishing scam detection on Ethereum: Towards financial security for blockchain ecosystem. In Proceedings of the 29th International Joint Conference on Artificial Intelligence (IJCAI) (pp. 4506–4512). Yokohama, Japan.
Chen, W., Zheng, Z., Cui, J., Ngai, E., Zheng, P., & Zhou, Y. (2018). Detecting Ponzi schemes on Ethereum: Towards healthier blockchain technology. In Proceedings of the World Wide Web Conference (pp. 1409–1418). Lyon, France.
Cover, T., & Hart, P. (1967). Nearest neighbor pattern classification. IEEE Transactions on Information Theory, 13(1), 21–27.
Dai, H., et al. (2018). Adversarial attack on graph structured data. In Proceedings of the International Conference on Machine Learning (ICML) (pp. 1115–1124).
Farrugia, S., Ellul, J., & Azzopardi, G. (2020). Detection of illicit accounts over the Ethereum blockchain. Expert Systems with Applications, 150.
Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 29(5), 1189–1232.
Huang, T., Lin, D., & Wu, J. (2022). Ethereum account classification based on graph convolutional network. IEEE Transactions on Circuits and Systems II, 69(5), 2528–2532.
Kleinbaum, D. G., Dietz, K., Gail, M., Klein, M., & Klein, M. (2002). Logistic regression. Springer.
Lin, D., Wu, J., Yuan, Q., & Zheng, Z. (2020a). Modeling and understanding Ethereum transaction records via a complex network approach. IEEE Transactions on Circuits and Systems II, 67(11), 2737–2741.
Lin, D., Wu, J., Yuan, Q., & Zheng, Z. (2020b). T-EDGE: Temporal weighted multidigraph embedding for Ethereum transaction network analysis. Frontiers in Physics, 8.
Liu, J., Zheng, J., Wu, J., & Zheng, Z. (2022). FA-GNN: Filter and augment graph neural networks for account classification in Ethereum. IEEE Transactions on Network Science and Engineering, 9(4), 2579–2588.
Sun, L., et al. (2018). Adversarial attack and defense on graph data: A survey. arXiv preprint arXiv:1812.10528.
Wen, H., Fang, J., Wu, J., & Zheng, Z. (2021). Transaction-based hidden strategies against general phishing detection framework on Ethereum. In Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS) (pp. 1–5).
Wu, J., Liu, J., Zhao, Y., & Zheng, Z. (2021). Analysis of cryptocurrency transactions from a network perspective: An overview. Journal of Network and Computer Applications, 190.
Wu, J., et al. (2022). Who are the phishers? Phishing scam detection on Ethereum via network embedding. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 52(2), 1156–1166.
Xia, Y., Liu, J., & Wu, J. (2022). Phishing detection on Ethereum via attributed ego-graph embedding. IEEE Transactions on Circuits and Systems II, 69(5), 2538–2542.
Yuan, Q., Huang, B., Zhang, J., Wu, J., Zhang, H., & Zhang, X. (2020). Detecting phishing scams on Ethereum based on transaction records. In Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS) (pp. 1–5).
Zheng, Z., Dai, H.-N., & Wu, J. (2021). Blockchain intelligence: Methods, applications and challenges. Springer.
Zügner, D., Akbarnejad, A., & Günnemann, S. (2019). Adversarial attacks on neural networks for graph data. In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI) (pp. 2847–2856).
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
