A Conceptual Data Model for Trajectory Data Mining

Data mining has become very popular in the last years, and it is well known that data preprocessing is the most effort and time consuming step in the discovery process. In part, it is because database designers do not think about data mining during the conceptual design of a database, therefore data are not prepared for mining. This problem increases for spatio-temporal data generated by mobile devices, which involve both space and time. In this paper we propose a novel solution to reduce the gap between databases and data mining in the domain of trajectories of moving objects, aiming to reduce the effort for data preprocessing. We propose a general framework for modeling trajectory patterns during the conceptual design of a database. The proposed framework is a result of several works including different data mining case studies and experiments performed by the authors on trajectory data modeling and trajectory data mining. It has been validated with a data mining query language implemented in PostGIS, that allows the user to create, instantiate and query trajectory data and trajectory patterns.