Evaluating Graph Neural Networks Models to Detect Fraudulent Users in e-Commerce

• Larissa de Andrade Silva UFAM http://orcid.org/0000-0001-9489-1729
• Eduardo L. Feitosa UFAM https://orcid.org/0000-0001-6401-3992

DOI: https://doi.org/10.5753/sbseg_estendido.2021.17361

Abstract

Models based on graphs and the combination with deep learning, the Graph Neural Networks (GNN), have been used to detect fraud in electronic commerce with promising results. In this paper, models that use node classification, based on neighborhood information, and community recognition, using real datasets were evaluated. The results, although promising (accuracy ranging from 50% to 86%), show that it is still necessary to study and investigate them better so that they can, in the future, act in anti-fraud solutions.

Keywords: gnn, e-commerce, fraud

References

ABCOMM (2020). O comércio eletrônico deve crescer 18% em 2020 e movimentar r$ 106 bilhões. https://bityli.com/3DjEH.

Hu, B., Zhang, Z., Shi, C., Zhou, J., Li, X., and Qi, Y. (2019). Cash-out user detection based on attributed heterogeneous information network with a hierarchical attention mechanism. Proceedings of the AAAI Conference on Articial Intelligence, 33(01):946–953.

Kipf, T. N. and Welling, M. (2017). Semi-supervised classication with graph convolutional networks. In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26. OpenReview.net.

Li, A., Qin, Z., Liu, R., Yang, Y., and Li, D. (2019). Spam review detection with graph convolutional networks. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management. ACM.

LIU, Z., CHEN, C., YANG, X., ZHOU, Junand LI, X., and SONG, L. (2018). Heterogeneous graph neural networks for malicious account detection. In 27th ACM International Conference On Information And Knowledge Management, pages 2077–2085.

Liu, Z., Dou, Y., Yu, P. S., Deng, Y., and Peng, H. (2020). Alleviating the inconsistency problem of applying graph neural network to fraud detection. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’20, page 1569–1572, New York, NY, USA. ACM.

Scarselli, F., Gori, M., Tsoi, A. C., Hagenbuchner, M., and Monfardini, G. (2009). The graph neural network model. IEEE Transactions on Neural Networks, 20(1):61–80.

Velickovíc, P., Cucurull, G., Casanova, A., Romero, A., Liò, P., and Bengio, Y. (2018). Graph attention networks. In 6th International Conference on Learning Representations, ICLR 2018, Conference Track Proceedings. OpenReview.net.

Wang, J., Wen, R., Wu, C., Huang, Y., and Xion, J. (2019). Fdgars: Fraudster detection via graph convolutional networks in online app review system. In Companion Proceedings of The 2019 World Wide Web Conference, WWW ’19, page 310–316, New York, NY, USA. ACM.

Wardhani, N. W. S., Rochayani, M. Y., Iriany, A., Sulistyono, A. D., and Lestantyo, P. (2019). Cross-validation metrics for evaluating classication performance on imbalanced data. In 2019 International Conference on Computer, Control, Informatics and its Applications (IC3INA), pages 14–18.

Zeng, Y. and Tang, J. (2021). Rlc-gnn: An improved deep architecture for spatial-based graph neural network with application to fraud detection. Applied Sciences, 11(12).

Zhang, Y., Fan, Y., Ye, Y., Zhao, L., and Shi, C. (2019). Key player identication in underground forums over attributed heterogeneous information network embedding framework. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management, CIKM ’19, page 549–558, New York, NY, USA. ACM.