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

Graph anomaly detection (GAD) is a vital task in graph-based machine learningand has been widely applied in many real-world applications. The primary goalof GAD is to capture anomalous nodes from graph datasets, which evidentlydeviate from the majority of nodes. Recent methods have paid attention tovarious scales of contrastive strategies for GAD, i.e., node-subgraph andnode-node contrasts. However, they neglect the subgraph-subgraph comparisoninformation which the normal and abnormal subgraph pairs behave differently interms of embeddings and structures in GAD, resulting in sub-optimal taskperformance. In this paper, we fulfill the above idea in the proposedmulti-view multi-scale contrastive learning framework with subgraph-subgraphcontrast for the first practice. To be specific, we regard the original inputgraph as the first view and generate the second view by graph augmentation withedge modifications. With the guidance of maximizing the similarity of thesubgraph pairs, the proposed subgraph-subgraph contrast contributes to morerobust subgraph embeddings despite of the structure variation. Moreover, theintroduced subgraph-subgraph contrast cooperates well with the widely-adoptednode-subgraph and node-node contrastive counterparts for mutual GAD performancepromotions. Besides, we also conduct sufficient experiments to investigate theimpact of different graph augmentation approaches on detection performance. Thecomprehensive experimental results well demonstrate the superiority of ourmethod compared with the state-of-the-art approaches and the effectiveness ofthe multi-view subgraph pair contrastive strategy for the GAD task.

Graph Anomaly Detection via Multi-Scale Contrastive Learning Networks with Augmented View