A model for enriching trajectories with semantic geographical information

This paper proposes a data preprocessing model to add semantic information to trajectories in order to facilitate trajectory data analysis in different application domains and shows that the query complexity for the semantic analysis of trajectories will be significantly reduced.

Abstract

The collection of moving ob ject data is becoming more and 
\nmore common, and therefore there is an increasing need 
\nfor the efficient analysis and knowledge extraction of these 
\ndata in different application domains. Tra jectory data are 
\nnormally available as sample points, and do not carry se- 
\nmantic information, which is of fundamental importance for 
\nthe comprehension of these data. Therefore, the analysis 
\nof tra jectory data becomes expensive from a computational 
\npoint of view and complex from a user’s perspective. En- 
\nriching tra jectories with semantic geographical information 
\nmay simplify queries, analysis, and mining of moving ob- 
\nject data. In this paper we propose a data preprocessing 
\nmodel to add semantic information to tra jectories in order 
\nto facilitate tra jectory data analysis in different application 
\ndomains. The model is generic enough to represent the im- 
\nportant parts of tra jectories that are relevant to the appli- 
\ncation, not being restricted to one specific application. We 
\npresent an algorithm to compute the important parts and 
\nshow that the query complexity for the semantic analysis of 
\ntra jectories will be significantly reduced with the proposed 
\nmodel.