Anomaly detection performance (AUROC).

<div><p>Anomaly detection in attributed networks is critical for identifying threats such as financial fraud and intrusions across social, e-commerce, and cyber-physical domains. Existing graph-based methods face two limitations: (i) embedding-based approaches obscure fine-grained struct...

詳細記述

保存先:

書誌詳細
第一著者:	Hossein Rafieizadeh (22676722) (author)
その他の著者:	Hadi Zare (20073000) (author), Mohsen Ghassemi Parsa (22676725) (author), Hocine Cherifi (8177628) (author)
出版事項:	2025
主題:	Cell Biology Science Policy Environmental Sciences not elsewhere classified Biological Sciences not elsewhere classified Information Systems not elsewhere classified intrusions across social reconstructions across views level contrastive learning dual contrastive learning across six benchmarks div >< p view discrepancies underutilized augmented graph views dcor improves auroc view discrepancies level contrast dual autoencoder augmented view specific information six datasets reduces auroc publicly available preserves fine physical domains performing non maximum gain leaving cross identifying threats financial fraud existing graph dcor reconstructs dcor ), contrasts reconstructions attributed networks attribute patterns
タグ:	タグ追加タグなし, このレコードへの初めてのタグを付けませんか!

Anomaly detection performance (AUROC).

類似資料