Data-Driven Attack Anomaly Detection in Public Transport Networks
Rui Yin, Nicholas Heng Loong Wong, Huaqun Guo, Wang Ling Goh · 2019
We present a method for attack detection in public transport networks. Through unsupervised machine learning, the daily data of the transportation system is clustered and a training model is established. Improved accuracy is achieved through self-organizing mapping and ensemble learning. We then apply the clustering model to assess the performance of the attack anomaly detection.