Towards Anomaly Detection on Relational Data

2026-06-16

AI Summary

This work addresses the challenge of anomaly detection in relational data, where high-dimensional heterogeneous attributes and complex cross-table join patterns are difficult to model. To this end, the authors propose RelAD, a framework that jointly models attribute reconstruction and relational edge reconstruction to effectively capture both semantic and structural anomaly signals. The method introduces a novel conditional sparse gating mechanism for attribute-level adaptive reconstruction and incorporates a dual-view multi-relational edge reconstruction module to integrate inter-table dependencies. Evaluated on six newly constructed benchmark datasets, RelAD significantly outperforms existing approaches, achieving superior detection performance while maintaining high computational efficiency.

Abstract

Relational databases are widely used for managing structured data in real-world systems. Detecting anomalies from such relational data is crucial for identifying fraud, risks, and abnormal behaviors, yet remains under-explored. The key challenges lie in the intrinsic complexity of relational data: multi-table attributes are high-dimensional and heterogeneous, making sparse abnormal clues easy to overwhelm by normal or irrelevant information; and anomalies may further manifest as abnormal connection patterns across different foreign-key relations, which existing tabular and graph anomaly detection methods are ill-suited to capture. To address them, we propose RelAD, a reconstruction-based framework that captures anomalies from both attribute and relational edge reconstruction. RelAD contains two core modules: conditional sparse-gated attribute reconstruction, which suppresses redundant multi-table attributes and emphasizes abnormal semantic blocks, and dual-view multi-relational edge reconstruction, which detects relation-specific abnormal connections from both intrinsic and behavioral entity profiles. The resulting attribute and relational signals are integrated through a lightweight fusion module to produce the final anomaly score. We further construct 6 benchmark datasets with systematic anomalies, on which extensive experiments show that RelAD consistently outperforms other baselines while achieving competitive efficiency.

Problem

• anomaly detection
• relational data
• foreign-key relations
• heterogeneous attributes
• abnormal connections

Innovation

• relational anomaly detection
• attribute reconstruction
• multi-relational edge reconstruction
• conditional sparse gating
• dual-view entity profiling