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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Fraud detection aims to discover fraudsters deceiving other users by, forexample, leaving fake reviews or making abnormal transactions. Graph-basedfraud detection methods consider this task as a classification problem with twoclasses: frauds or normal. We address this problem using Graph Neural Networks(GNNs) by proposing a dynamic relation-attentive aggregation mechanism. Basedon the observation that many real-world graphs include different types ofrelations, we propose to learn a node representation per relation and aggregatethe node representations using a learnable attention function that assigns adifferent attention coefficient to each relation. Furthermore, we combine thenode representations from different layers to consider both the local andglobal structures of a target node, which is beneficial to improving theperformance of fraud detection on graphs with heterophily. By employing dynamicgraph attention in all the aggregation processes, our method adaptivelycomputes the attention coefficients for each node. Experimental results showthat our method, DRAG, outperforms state-of-the-art fraud detection methods onreal-world benchmark datasets.

Dynamic Relation-Attentive Graph Neural Networks for Fraud Detection