Attention-augmented and depthwise separable convolutional message passing for robust fraud detection in large-scale graphs

Ijeoma Amuche Chikwendu, Xiaoling Zhang, Chiagoziem C. Ukwuoma, Chibueze D. Ukwuoma, Isaac Adjei-Mensah, Chinedu I. Otuka, Chukwuebuka Joseph Ejiyi, Yeong Hyeon Gu, Mugahed A. Al–antariJournal of Advanced Research 2025 · DOI

Abstract

By actively learning from both homophilic and heterophilic neighbours and optimizing message passing with efficient convolutions and attention, ADSCMP enhances both the accuracy and scalability of fraud detection.
BibTeX
@article{chikwendu2025attention,
  author = {Ijeoma Amuche Chikwendu and Xiaoling Zhang and Chiagoziem C. Ukwuoma and Chibueze D. Ukwuoma and Isaac Adjei-Mensah and Chinedu I. Otuka and Chukwuebuka Joseph Ejiyi and Yeong Hyeon Gu and Mugahed A. Al–antari},
  title = {{Attention-augmented and depthwise separable convolutional message passing for robust fraud detection in large-scale graphs}},
  journal = {Journal of Advanced Research},
  year = {2025},
  doi = {10.1016/j.jare.2025.06.087},
}

Loading related content...

Figure