An ontological approach to spatio-temporal information modelling in transportation

Nowadays there is too much data available. This is because almost every electronic device around us generates data. Usage of this resource is a must if one want to improve transportation. Having roads a variety of sensors installed, their data readability, comprehension and usage is of vital importance for a good traffic analysis. Data integration is needed to analyse information from these different sensors. Having this in mind, it is also known that this task is difficult if done manually, necessitating appropriate research and new techniques that can simplify such a task and preferentially automate it. The concept of ontology is particularly suitable to address data integration issue while maintaining their original meaning and representation. This paper explores the ontology concept so as to leverage data integration through a service-oriented architecture envisaging applications in transportation systems. Such a perspective allows potential clients to easily access heterogeneous data accounting neither for their original schema nor for their location and origin. To illustrate the approach, a prototype was implemented, which uses data from GPS logs, network topology from OpenStreetMap and counts from inductive loops.