Extracting Mobility Statistics from Indexed Spatio-Temporal Datasets

With the recent progress of spatial information technologies and mobile computing technologies, spatio-temporal databases which store information on moving objects including vehicles and mobile users have gained a lot of research interests. In this paper, we propose an algorithm to extract mobility statistics from indexed spatio-temporal datasets for the interactive analysis of huge collections of moving object trajectories. We focus on a mobility statistics value called the Markov transition probability, which is based on a cell-based organization of a target space and the Markov chain model. The proposed algorithm efficiently computes the specified Markov transition probabilities with the help of a spatial index R-tree. We reduce the statistics computation task to a kind of constraint satisfaction problem that uses a spatial index, and utilize internal representation of R-tree in an efficient manner. Key words spatio-temporal databases, mobility statistics, Markov chain model, mob...

• Type: Conference
• Year: 2004
• Where: STDBM
• Authors: Yoshiharu Ishikawa, Yuichi Tsukamoto, Hiroyuki Kitagawa
• Keywords: Database, Markov Chain Model, Markov Transition, Mobility Statistics, STDBM 2004