<ul><li>A new framework, Dual-channel Heterophilic Message Passing (DHMP), is proposed for fraud detection.</li><li>DHMP leverages a heterophily separation module to divide the graph into homophilic and heterophilic subgraphs.</li><li>It applies shared weights to capture signals at different frequencies independently and incorporates a customized sampling strategy for training.</li><li>Extensive experiments demonstrate that DHMP outperforms existing methods for fraud detection.</li></ul>

Dual-channel Heterophilic Message Passing for Graph Fraud Detection

Discover more