Data-Driven Anomaly Detection and Prediction in Public Transportation Networks

Anomaly detection and prediction typically require extensive domain knowledge to develop tools capable of automatically identifying or forecasting anomalous events or behaviors in IoT systems. These systems—particularly those within Public Transportation Networks—often consist of devices with diverse capabilities, functions, and operational lifespans, making the detection of rare anomalies especially challenging. Moreover, establishing domain expertise and collecting a sufficient number of data points for anomaly detection is frequently time-consuming and <a href="http://costly.

Virtual:” target=”_blank” title=”costly.

Virtual:”>costly.

Virtual: https://events.vtools.ieee.org/m/481954