Identifying the Hotspots in Urban Areas Using Taxi GPS Trajectories

Taxi GPS trajectories contain massive spatial and temporal information of urban functions and residents' activities. The mining of hotspots and moving patterns of residents from taxi trajectory data is of great significance for urban planning, traffic management, public travel services, etc. Because of the existing hotspots region mining methods have low computational efficiency in the face of large-scale trajectories data, we build a directed weighted network to model intra-city spatial interactions and use the DWNodeRank algorithm to reveal hotspots. Firstly, the urban area is divided into grid cells, and the origin-destination matrix is produced by travel records between grid cells. Secondly, a directed weighted network is built based on origin-destination matrix. Finally, the taxi trajectory data of Lanzhou city in China are taken as the experimental data. From the result, we find the distribution and strength of hotspots area in Lanzhou city.