To address this, researchers proposed a novel anomaly detection model called the Contextual Auto-Encoder (CAE), which leverages machine learning to identify irregularities in flight trajectories[2].