On-line location prediction exploiting spatial and velocity context

We treat the problem of movement prediction as a classification task. We assume the existence of a (gradually populated and/or trained) knowledge base and try to compare the movement pattern of a certain object with stored information in order to predict its future location. We introduce a novel distance metric function based on weighted spatial and velocity context used for location prediction. The proposed distance metric is compared with other distance metrics in the literature on real traffic data and reveals its superiority.