An Explainable GNN Framework for Component-Level Anomaly Diagnosis
An explainable GNN-based anomaly detection framework is proposed that shifts the perspective from sensor-level anomalies to component-level diagnosis, hypothesizing that anomalous measurements are symptoms of altered inter-sensor influences.
Sena Ozgunay, Louise Travé-Massuyès, Jean-Michel Loubes et al.
· 0 citations