Detecting Congestion and Detour of Taxi Trip via GPS Data

With the continuous development of urbanization, taxi has been one of the main transports in daily trip. The current study of the taxi trip is more about the traffic situation and appropriate demand in this city. Our study by using DBSCAN (Density-Based Spatial Clustering of Applications with Noise), and find the abnormal driving items via taxi GPS data. Then we proposed a method to detect the congestion and detour that result in theses anomaly. We hope our work can provide useful references for the relevant departments to manage urban traffic.