Selecting a representation for spatial vagueness: A decision making approach

Abstract

Representing vague places is a challenge in information systems. There are several approaches, each differing in aspects such as the underlying assumptions they make about space, their data models and reasoning abilities. Despite this there is no general solution and the question of which method to select is a matter of fitness for purpose. So far no methodology exists to support choosing the appropriate representation for a given problem. A formal decision making approach is presented here to select a suitable modelling technique to represent vague places. To do this, the criteria on the basis of which the decision is made are derived first. Commonly used methods to model spatial vagueness and uncertainty are then analyzed on the basis of these criteria. Finally, we describe a methodology that uses the analytic hierarchy process, in order to provide a quantitative ranking of candidate methods in their order of suitability for an application scenario.