Accepted Unknow ; Published 2024 December 1

Journal of Industrial Engineering and Applied Science, 2024, 2(6), 3005-6071.

View PDFDownload PDFArchive (OJS)

This paper proposes a novel multi-modal graph neural network framework for detecting market manipulation in high-frequency trading environments. The framework integrates diverse data sources through sophisticated fusion mechanisms and employs attention-based graph neural networks to capture complex trading patterns. Our approach constructs dynamic trading networks that encode temporal and structural dependencies, enabling the detection of subtle manipulation strategies. The model architecture incorporates multiple attention layers for feature selection and cross-modal information fusion, achieving superior detection performance compared to traditional methods. Experimental results on real-world high-frequency trading data from major exchanges demonstrate the framework's effectiveness, reaching 98.7% accuracy in manipulation detection while maintaining low latency (8.3ms average processing time). The model exhibits robust performance across various market conditions and manipulation patterns, with precision and recall rates exceeding 97%. Through comprehensive case studies and interpretability analysis, we validate the framework's ability to identify and explain complex manipulation strategies while providing insights for regulatory compliance. The proposed approach advances state-of-the-art market surveillance technology, offering a scalable solution for real-time manipulation detection in modern financial markets.

Market Manipulation Detection , Graph Neural Networks , Multi-modal Data Fusion , High-Frequency Trading .

APA Style

Chen, Y., Li, M., Shu, M., Bi, W. & Xia, S. (2024). Multi-modal market manipulation detection in high-frequency trading using graph neural networks. Journal of Industrial Engineering and Applied Science, 2(6), 111-120. https://doi.org/10.70393/6a69656173.323432

I want to extend my sincere gratitude to Yitian Zhang, Yibang Liu, and Shuaiqi Zheng for their groundbreaking research on graph neural network approaches in fraud detection, as published in their article "A Graph Neural Network-Based Approach for Detecting Fraudulent Small-Value High-Frequency Accounting Transactions"[23]. Their innovative methodology and comprehensive analysis have significantly influenced my understanding of graph-based deep learning applications in financial fraud detection and provided valuable insights for my research. I would also like to express my appreciation to Siwei Xia, Yida Zhu, Shuaiqi Zheng, Tianyi Lu, and Ke Xiong for their pioneering work on deep learning applications in financial risk assessment, as published in their article "A Deep Learning-based Model for P2P Microloan Default Risk Prediction"[24]. Their systematic approach to model development and performance evaluation has dramatically enhanced my knowledge of deep learning architectures in financial applications and inspired significant aspects of this research.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.