English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Graph neural networks (GNN) have emerged as a powerful tool for frauddetection tasks, where fraudulent nodes are identified by aggregating neighborinformation via different relations. To get around such detection, craftyfraudsters resort to camouflage via connecting to legitimate users (i.e.,relation camouflage) or providing seemingly legitimate feedbacks (i.e., featurecamouflage). A wide-spread solution reinforces the GNN aggregation process withneighbor selectors according to original node features. This method may carrylimitations when identifying fraudsters not only with the relation camouflage,but with the feature camouflage making them hard to distinguish from theirlegitimate neighbors. In this paper, we propose a Hierarchical Attention-basedGraph Neural Network (HA-GNN) for fraud detection, which incorporates weightedadjacency matrices across different relations against camouflage. This ismotivated in the Relational Density Theory and is exploited for forming ahierarchical attention-based graph neural network. Specifically, we design arelation attention module to reflect the tie strength between two nodes, whilea neighborhood attention module to capture the long-range structural affinityassociated with the graph. We generate node embeddings by aggregatinginformation from local/long-range structures and original node features.Experiments on three real-world datasets demonstrate the effectiveness of ourmodel over the state-of-the-arts.

Improving Fraud Detection via Hierarchical Attention-based Graph Neural Network