A Trajectory Ontology Design Pattern for Semantic Trajectory Data Warehouses: Behavior Analysis and Animal Tracking Case Studies

A Trajectory Ontology Design Pattern for Semantic Trajectory Data Warehouses: Behavior Analysis and Animal Tracking Case Studies

Marwa Manaa (University of Tunis, Tunisia), Thouraya Sakouhi (University of Tunis, Tunisia) and Jalel Akaichi (University of Bisha, Saudi Arabia)
Copyright: © 2019 |Pages: 22
DOI: 10.4018/978-1-5225-5516-2.ch004


Mobility data became an important paradigm for computing performed in various areas. Mobility data is considered as a core revealing the trace of mobile objects displacements. While each area presents a different optic of trajectory, they aim to support mobility data with domain knowledge. Semantic annotations may offer a common model for trajectories. Ontology design patterns seem to be promising solutions to define such trajectory related pattern. They appear more suitable for the annotation of multiperspective data than the only use of ontologies. The trajectory ontology design pattern will be used as a semantic layer for trajectory data warehouses for the sake of analyzing instantaneous behaviors conducted by mobile entities. In this chapter, the authors propose a semantic approach for the semantic modeling of trajectory and trajectory data warehouses based on a trajectory ontology design pattern. They validate the proposal through real case studies dealing with behavior analysis and animal tracking case studies.
Chapter Preview


Advances in pervasive systems triggered by the incredible technical evolution of mobile devices and positioning technologies led to the eruption of disparate, dynamic, and geographically distributed mobility data. For a long while, location sensing devices and wireless networks started becoming widely untethered (Yan & Chakraborty., 2007). As a result, disparate mobility data revealing the details of instantaneous activities conducted by mobile entities can be collected and used for any mobile object trajectory reconstruction.

Note that, trajectory data, which is a record set of gathered mobility data, can be associated to different domain-specific information. Trajectories are naturally represented as raw trajectory denoting a sequence of temporally-indexed positions. For example, pedestrian displacement is described using a time-varying point which is a point whose position evolves over the time. In other cases, such as studying bird migration displacement, trajectories are defined by decision spatio-temporal points i.e., stops and moves according to predefined paths i.e., sub-trajectories. We will refer to the latter cases as structured trajectory (Spaccapietra et al., 2008). In other cases, trajectory with Region Of Interest (ROI) (Giannotti et al., 2007) represents trajectory data as a sequence of regions and time intervals. The phenomenon of adopt-ing raw trajectory referencing domain ontologies by organizations generates a new type of trajectory, called semantic trajectory. Semantic trajectory (Alvares et al., 2007), (Bogorny et al., 2009), (Yan & Chakraborty., 2007), (Richter et al.,2015) and trajectory with Semantic ROI (Yan., 2009) annotates decision points with con-textual information and enrich them by links with geographic and application domain concepts. In other cases, space-time path (Wannous et al., 2013) extends semantic trajec-tories with mobile object activity performed during the travel. An example of such trajectories occurs in Location-Based Social Networks (LBSN), where the raw trajectory are user check-ins to Points Of Interest (POI) and the contextual information includes names of POI and activities during the travel.

In a highly heterogeneous and dynamic environment, such as the Web, arriving at commonly agreed and stable domain ontologies is a prone-to-fail task and progress has been slow over the last years (Hu et al., 2013). Ontology design patterns have emerged as more flexible, reusable and manageable modeling solutions (Gangemi., 2005). It may provide common model for different representations of trajectory data where designers can pick the appropriate knowledge to define trajectories in view of share, exchange or integration. Alongside, data warehousing techniques are expected to analyze and extract valuable information from heterogeneous trajectory data sources.

Key Terms in this Chapter

Model: An abstraction of a system designed as a set of facts constructed in a particular intention. It must be used to answer questions about the system under study. In the field of trajectory data, models are used to formalize and analyze trajectories of mobile objects.

Trajectory: Trajectory is the record of a time-varying spatial phenomenon. Trajectory consists in the description of the movement of some moving objects at specific moment’s time. In reality, trajectory has to be built from a set of sample points which correspond to moving object positions.

Ontology: Ontology is a cognitive artefact allowing the shared design and operation for knowledge. Ontology is composed from concepts related to a domain of interest linked with relations.

Ontology Design Pattern: Is a modelling solution to solve recurrent ontology development problems that can be solved by means of a set of rules and shared guidelines that are packaged in the form of pattern.

Mobile Object: It is an identifiable geometries real word element that moves. Geometries may be points, lines, areas, or volumes, changing over the time like person, car, or natural phenomenon.

Reasoning Mechanism: Reasoning mechanisms allow deriving new facts from existing concepts and roles that are not expressed in the initial ontology.

Conceptual Model: A high-level description for a system. It allows us understanding and interpreting information related to a field. The later formalize information using a language in order to construct a system. The conceptual model includes a graphical representation of model concepts and represents relations between these elements.

Complete Chapter List

Search this Book: