ABSTRACT
Numerous techniques have been proposed for trajectory modeling (Ashbrook and Starner 2003, Yu et al. 2009) and for mobility data extraction (Giannotti et al. 2007, Nicholson and Noble 2008, Yavas et al. 2005). Most of these state-of-the-art techniques are novel. However, they focus only on the spatiotemporal dimensions to extract or build a trajectory model from raw geospatial data. Social network activities (e.g., Twitter messages, Facebook activities), telephone calls, or disease attributes are
not considered for trajectory modeling even if they are included in the raw mobility data. Big geospatial data emanating from smart phones and other state-of-the-art ubiquitous devices have a finite but wide variety of dimensions. Since existing trajectory models (Ashbrook and Starner 2003, Giannotti et al. 2007, Yu et al. 2009) are unable to handle or utilize these non-spatiotemporal dimensions, we strongly believe a new trajectory model that incorporates and processes dimensions beyond space and time is required to meet the demands of this rich diverse multidimensional big geospatial data. Due to this observation, we propose a novel trajectory model called Time Mobility Context Correlation Pattern (TMC-Pattern).
