Human mobility prediction based on individual and collective geographical preferences

Abstract

Understanding and predicting human mobility is a crucial component of transportation planning and management. In this paper we propose a new model to predict the location of a person over time based on individual and collective behaviors. The model is based on the person's past trajectory and the geographical features of the area where the collectivity moves, both in terms of land use, points of interests and distance of trips. The effectiveness of the proposed prediction model is tested using a massive mobile phone location dataset available for the Boston metropolitan area. Experimental results show good levels of accuracy in terms of prediction error and prove the advantage of using the collective behavior in the prediction model.