Home // GEOProcessing 2013, The Fifth International Conference on Advanced Geographic Information Systems, Applications, and Services // View article

Trajectory Data Mining: a Novel Distance Measure

Authors:
Ayman Al-Serafi
Ahmed Elragal

Keywords: distance measurement; geospatial data analysis; moving-object databases; spatiotemporal data; trajectory data mining

Abstract:
There is currently an increasing availability of large spatiotemporal datasets. Sequences of spatiotemporal data or paths, also known as trajectories, can be captured by modern technology and stored in moving-object databases (MOD) or a trajectory data warehouse. It is a common challenge for knowledge discovery within MODs to query proximities and distances, e.g. in clustering trajectories. Previously adopted distance measures focus on the complexity of geometric and/or mathematical models of trajectories, while ignoring several aspects common to all spatiotemporal trajectories, e.g. direction, distance covered, and duration. This research introduces a more comprehensive approach for trajectory distance measurement in spatiotemporal applications. The approach is simplified, yet novel, introducing a new set of dimensional variables, therefore called the Multi-Dimensional Trajectory Distance Measure (MTDM). The accuracy and relevance of MTDM is evaluated in experiments using multiple proximity metrics, for example MTDM based on Euclidean proximity calculation. A geospatial data analysis framework is utilized in the experiments. Efficiency evaluation of MTDM showed the feasibility of applying the measure to various trajectory datasets.

Pages: 125 to 132

Copyright: Copyright (c) IARIA, 2013

Publication date: February 24, 2013

Published in: conference

ISSN: 2308-393X

ISBN: 978-1-61208-251-6

Location: Nice, France

Dates: from February 24, 2013 to March 1, 2013