Hybrid Tabular-Graph Representation Learning Approach for Improving Fraud Detection in Ethereum Network

doi:10.23940/ijpe.26.11.p6.681693

Abstract

Abstract: The Ethereum transaction network contains legitimate transactions as well as fraudulent transactions related to phishing, fraud, and other scams. Although traditional detection models rely on transaction-level features, relationships between senders, receivers and transactions may provide additional information for the detection of malicious intent. However, controlled evaluations comparing tabular and graph-based features in binary and multiclass classifications are still limited. To this end, this paper presents a hybrid tabular-graph framework for fraud detection at the Ethereum transaction level. The proposed methodology was assessed against 71,249 transactions recorded between October 2017 and December 2019, including 57,000 regular transactions, 3,110 phishing transactions and 11,139 fraudulent transactions. From the sender and the recipient addresses, a heterogeneous client-transaction-merchant graph is generated. The structural representation is then extracted using Graphical Convolutional Networks (GCN), Graph SAGE, and Fast Inductive Graph Representation Learning (FI-GRL), and combined with selected tabular attributes. The ensemble classifiers such as Random Forest, XGBoost and LightGBM are used for classification. The proposed approach is assessed using a 5-fold stratified cross-validation without re-sampling for binary classification (normal vs. fraudulent) and multi-class classification (normal, phishing and fraud). The results show that the augmenting graph structural features improve the classification. The FI-GRL and LightGBM combinations provided the highest total performance, with Mathew Correlation Coefficient (MCC) of 0.9912 and 0.9894 respectively on binary and multiclass classification compared to LightGBM with MCC of 0.8373 and 0.8124 respectively for binary and multiclass classification with tabular features. These findings suggest that the integration of structured information derived from graphs, in addition to traditional table attributes, can significantly improve the detection of transaction level fraud in Ethereum network.

Key words: Ethereum fraud detection, graph embedding, FI-GRL, LightGBM, phishing, random forest, scamming, XGBoost

Vaibhav Ranjan and Potukuchi Raghu Vamsi. Hybrid Tabular-Graph Representation Learning Approach for Improving Fraud Detection in Ethereum Network [J]. Int J Performability Eng, 2026, 22(11): 681-694.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

References

[1] Ghnemat R., andMosa H., 2025. Blockchain-based fraud detection: A systematic review of Ethereum network applications.Cluster Computing, 28(16), 1080.
[2] Farrugia S., Ellul J., andAzzopardi G., 2020. Detection of illicit accounts over the Ethereum blockchain expert systems with applications,150.
[3] Kimber J., Branca E., Natadze A., Stakhanova N., Yao Y., Hou J., Yuan M., Zhang H., Xu Z., Li X.Y., andCernera F., 2025. Internet technology. ACM Transactions on,25(3).
[4] Li C.T., Tsai Y.C., andLiao J.C., 2023. Graph neural networks for tabular data learning. In2023 IEEE 39th International Conference on Data Engineering (ICDE), pp. 3589-3592.
[5] Aziz R.M., Baluch M.F., Patel S., andGanie A.H., 2022. LGBM: a machine learning approach for Ethereum fraud detection. International Journal of Information Technology,14(7), pp. 3321-3331.
[6] Biau G., andScornet E., 2016. A random forest guided tour. Test,25(2), pp. 197-227.
[7] Krishnapuram B., Shah M., Smola A., Aggarwal C., Shen D., Rastogi R., Chen T., andGuestrin C., 2016. XGBoost: a scalable tree boosting system. InProc 22nd ACM Sigkdd Int Conf Knowl Discov Data Min, pp. 785-794.
[8] Ke G., Meng Q., Finley T., Wang T., Chen W., Ma W., Ye Q., andLiu T.Y., 2017. Lightgbm: A highly efficient gradient boosting decision tree.Advances in Neural Information Processing Systems, 30.
[9] Gu Z., andDib O., 2025. Enhancing fraud detection in the Ethereum blockchain using ensemble learning.PeerJ Computer Science, 11, e2716.
[10] Ertam F.,2025. Near real-time Ethereum fraud detection using explainable AI in blockchain networks.Applied Sciences, 15(19), 10841.
[11] Chithanuru V., andRamaiah M., 2025. Proactive detection of anomalous behavior in Ethereum accounts using XAI-enabled ensemble stacking with Bayesian optimization.PeerJ Computer Science, 11, e2630.
[12] Tan R., Tan Q., Zhang P., andLi Z., 2021. Graph neural network for Ethereum fraud detection. In2021 IEEE International Conference on Big Knowledge (ICBK), pp. 78-85.
[13] Li P., Xie Y., Xu X., Zhou J., andXuan Q., 2022. Phishing fraud detection on Ethereum using graph neural network. InInternational Conference on Blockchain and Trustworthy Systems, pp. 362-375.
[14] Li S., Gou G., Liu C., Hou C., Li Z., andXiong G., 2022. TTAGN: temporal transaction aggregation graph network for Ethereum phishing scams detection. InProceedings of the ACM Web Conference 2022, pp. 661-669.
[15] Zhang W., andLuo C., 2025. Ge-gnn: gated edge-augmented graph neural network for fraud detection. IEEE Transactions on Big Data,11(4), pp. 1664-1676.
[16] Qiao H., Tong H., An B., King I., Aggarwal C., andPang G., 2025. Deep graph anomaly detection: A survey and new perspectives. IEEE Transactions on Knowledge and Data Engineering,37(9), pp. 5106-5126.
[17] Munoz M.F., Zhang K., andAmara F., 2022. ZipZap: a blockchain solution for local energy trading. In2022 IEEE International Conference on Blockchain and Cryptocurrency (ICBC), pp. 1-5.
[18] van Veen K., Ahmed F., andvan Keulen M., 2025. XAI in fraud detection: A causal perspective. InWorld Conference on Explainable Artificial Intelligence, pp. 317-329.
[19] Abbas Jasim Al‐Hchaimi A., Khalifa M.A., andEl‐Shafai W., 2026. Explainable AI with imbalanced learning strategies for blockchain transaction fraud detection.Engineering Reports, 8(1), e70545.
[20] Lundberg S.M., andLee S.I., 2017. A unified approach to interpreting model predictions.Advances in Neural Information Processing Systems, 30.
[21] Liao J.C., Chiu J.W., andLi C.T., 2025. TabGSL: graph structure learning for tabular data predictions. In2025 IEEE International Conference on Big Data (BigData), pp. 4155-4164.
[22] Zhou K., Liu Z., Chen R., Li L., Choi S.H., andHu X., 2022. Table2Graph: transforming tabular data to unified weighted graph. InIJCAI, pp. 2420-2426.
[23] Aziz R.M., Baluch M.F., Patel S., andKumar P., 2022. A machine learning based approach to detect the Ethereum fraud transactions with limited attributes. Karbala International Journal of Modern Science,8(2), pp. 139-151.
[24] Zhang W., Li R., Bai Q., andWang S., 2025. SparseFraudNet: A graph-based approach for cold-start fraud detection with information aggregation. ACM Transactions on Information Systems,44(1), pp. 1-45.
[25] Li C.T., Tsai Y.C., Chen C.Y., andLiao J.C., 2025. Graph neural networks for tabular data learning: A survey with taxonomy and directions. ACM Computing Surveys,58(1), pp. 1-51.
[26] Kipf T.N., andWelling M., 2016. Semi-supervised classification with graph convolutional networks.Arxiv Preprint Arxiv:1609.02907.
[27] Hamilton W., Ying Z., andLeskovec J., 2017. Inductive representation learning on large graphs.Advances in Neural Information Processing Systems, 30.
[28] Van Belle R., Van Damme C., Tytgat H., andDe Weerdt J., 2022. Inductive graph representation learning for fraud detection.Expert Systems with Applications, 193, 116463.
[29] Jiang F., Zheng L., Xu J., andYu P., 2018. Fi-grl: fast inductive graph representation learning via projection-cost preservation. In2018 IEEE International Conference on Data Mining (ICDM), pp. 1067-1072.
[30] Al-E’mari S., Anbar M., Sanjalawe Y., andManickam S., 2020. A labeled transactions-based dataset on the Ethereum network. InInternational Conference on Advances in Cyber Security, pp. 61-79.
[31] Naidu G., Zuva T., andSibanda E.M., 2023. A review of evaluation metrics in machine learning algorithms. InComputer Science On-line Conference, pp. 15-25.