Date of Award

12-2005

Level of Access Assigned by Author

Campus-Only Thesis

Degree Name

Master of Science (MS)

Department

Spatial Information Science and Engineering

Advisor

Kathleen Stewart Hornsby

Second Committee Member

Max J. Egenhofer

Third Committee Member

Mary Kate Beard-Tisdale

Abstract

The dynamic aspects of a geospatial domain can be modeled realistically only by extending the traditional static geospatial data model to a more comprehensive spatiotemporal model. These dynamic aspects include the actions, happenings and activities, referred to as occurrents, taking place in the domain as well as the changes to the 'things' or continuants that exist there. In this thesis, an event-based approach is used to model a prototypical dynamic geospatial domain - a harbor - where harbor and vesselrelated occurrents are modeled as events. This thesis shows how event-related semantics are vital for understanding, describing, and predicting change in geospatial domains. The higher-level semantics associated with a domain's important occurrents are modeled as noteworthy events, a key construct for enabling event notification services. This thesis formally defines three kinds of notification triggers: noteworthy events, noteworthy sequences, and noteworthy patterns, enabling the automation of event notifications in a spatiotemporal data model. Movement, for example the movement of a ship through a harbor, is perceived as a series of occurrents experienced by a continuant, providing the basis for modeling object movement as a sequence of events. Characteristic trends of an object's movement as well as the movement patterns of multiple moving objects are can be characterized by event patterns. Key patterns, like the repeating, collocating, and reiterating patterns in the harbor domain, are associated with unusual or unexpected movement trends and patterns.

Share