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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Graph-based Neural Networks (GNNs) are recent models created for learningrepresentations of nodes (and graphs), which have achieved promising resultswhen detecting patterns that occur in large-scale data relating differententities. Among these patterns, financial fraud stands out for itssocioeconomic relevance and for presenting particular challenges, such as theextreme imbalance between the positive (fraud) and negative (legitimatetransactions) classes, and the concept drift (i.e., statistical properties ofthe data change over time). Since GNNs are based on message propagation, therepresentation of a node is strongly impacted by its neighbors and by thenetwork's hubs, amplifying the imbalance effects. Recent works attempt to adaptundersampling and oversampling strategies for GNNs in order to mitigate thiseffect without, however, accounting for concept drift. In this work, we conductexperiments to evaluate existing techniques for detecting network fraud,considering the two previous challenges. For this, we use real data sets,complemented by synthetic data created from a new methodology introduced here.Based on this analysis, we propose a series of improvement points that shouldbe investigated in future research.

How effective are Graph Neural Networks in Fraud Detection for Network Data?