City Research Online

Abstract

The manipulation of geographic information (GI) has been considered, by most of its users, as a very complex process and has involved the development of specific applications called geographic information systems (GIS). A new subdiscipline of Information Science called Geographic Information Science (GISc) has been proposed by the GIS research community on the grounds that there are specific characteristics not only to GI but also to the processes involved in its manipulation.

The thesis draws on several research issues which are part of the agenda in GISc: The complexity of handling spatial/geographic information, spatial reasoning in dynamical systems, integration of several types of application, human-computer interaction and spatio-temporal issues.

In this context, this dissertation proposes the application of a new computational paradigm, intelligent agents in GISc. Intelligent agents are “computational systems that inhabit some complex dynamic environment, sense and act autonomously, and by doing so realise a set of goals or tasks for which they are designed “ (Maes, 1995).

The aim of this dissertation is to analyse the potential of research in intelligent agents in GISc and to explore the use of simple learning techniques to improve the adaptability of spatial intelligent agents. The thesis involves the following objectives: to analyse the needs of research in GISc in the areas of reasoning about geographic space; to study the potential of intelligent agents in that area of research; to explore the use of simple learning techniques to improve the adaptability of intelligent agents for geographic information; and to explore the implementation environments of GIS software for the integration of intelligent agent systems.

The primary contributions of this research are three case studies which use intelligent agents in a spatial or geographic context: a simple non-adaptive interface assistant for the printing and plotting tool of Smallworld GIS; an intelligent assistant that uses memory-based reasoning to identify and locate specific-purpose geographic information; a simulation of a car park where agents are cars that use reinforcement learning techniques to improve their parking performance.

Publication Type:	Thesis (Doctoral)
Subjects:	G Geography. Anthropology. Recreation > G Geography (General) Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science
Departments:	School of Communication & Creativity > Media, Culture & Creative Industries > Library & Information Science School of Communication & Creativity > School of Communication & Creativity Doctoral Theses Doctoral Theses

Export

Downloads