ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Download
Citation
Articles | Volume II-4/W2
https://doi.org/10.5194/isprsannals-II-4-W2-143-2015
https://doi.org/10.5194/isprsannals-II-4-W2-143-2015
10 Jul 2015
 | 10 Jul 2015

CLUSTERING ANALYSIS OF OFFICER'S BEHAVIOURS IN LONDON POLICE FOOT PATROL ACTIVITIES

J. Shen and T. Cheng

Keywords: Clustering, Behaviour Pattern, Similarity, GPS Trajectory, Region of Interest

Abstract. In this small paper we aim at presenting a framework of conceptual representation and clustering analysis of police officers’ patrol pattern obtained from mining their raw movement trajectory data. This have been achieved by a model developed to accounts for the spatio-temporal dynamics human movements by incorporating both the behaviour features of the travellers and the semantic meaning of the environment they are moving in. Hence, the similarity metric of traveller behaviours is jointly defined according to the stay time allocation in each Spatio-temporal region of interests (ST-ROI) to support clustering analysis of patrol behaviours. The proposed framework enables the analysis of behaviour and preferences on higher level based on raw moment trajectories. The model is firstly applied to police patrol data provided by the Metropolitan Police and will be tested by other type of dataset afterwards.