A Grid-Based and a Context-Oriented Trajectory Modeling for Mobility Prediction in Smart Cities

In the last decade, mobility prediction has played a crucial role in urban planning, traffic forecasting, advertising, and service recommendation. This paper addresses the prediction of mobility and emphasizes an essential step that is trajectory modeling (better the modelling is, better is the prediction). First, we propose a context-based and prediction-oriented trajectory model. Our model is based on a grid-oriented trajectory description technique that allows overcoming low precision and ambiguity issues. Second, our model is compared to some related trajectory models. Third, an application of the model in intelligent transportation domain is illustrated. Finally, to evaluate our model, we experiment it on a data mining-based prediction algorithm and show the results in terms of prediction accuracy.