Anomalies in attributed networks.

<p>Structural anomaly: an unexpected inter-community bridge, where the orange node sits in the cut between two communities and forms a shortcut (dashed) edge to the right cluster. Attribute anomaly: the orange node’s feature vector (colored bars) deviates from those of its neighbors even thoug...

وصف كامل

محفوظ في:

التفاصيل البيبلوغرافية
المؤلف الرئيسي:	Hossein Rafieizadeh (22676722) (author)
مؤلفون آخرون:	Hadi Zare (20073000) (author), Mohsen Ghassemi Parsa (22676725) (author), Hocine Cherifi (8177628) (author)
منشور في:	2025
الموضوعات:	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
الوسوم:	إضافة وسم لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!

الوصف
الملخص:	<p>Structural anomaly: an unexpected inter-community bridge, where the orange node sits in the cut between two communities and forms a shortcut (dashed) edge to the right cluster. Attribute anomaly: the orange node’s feature vector (colored bars) deviates from those of its neighbors even though its connectivity looks normal. Interaction anomaly: a structure–attribute mismatch, where the orange node’s attributes align with the left community while its links embed it in the right community. Visual encoding: blue = normal nodes and edges; orange = anomalous node or edge; dashed orange edge = anomalous link; colored bars = node attributes.</p>

Anomalies in attributed networks.

مواد مشابهة