ARTIFICIAL INTELLIGENCE,SPATIAL ANALYSIS, AND SPATIAL DECISION MAKING
Yang Leang(Department of Geography and Centre for Environmental Studies, the Chinese University of Hong Kong, Hong Kong)
AbsractThis paper gives a brief analysis of the development and applications of artificial intelli-gence in spatial analysis and spatial decision making. Basic approaches to the representation of and inference with unstructured spatial knowledge are first introduced. They include preposi-tional and predicate logic, production systems,semantic networks, frames, and object-orented systems. The paper then discusses uncertainty in spatial inference. Basic frameworks such as probabilistic inference, subjective Bayesian,certainty factor, evidence theory,and fuzzy infer-ence are briefly examined. Theory of inclusion, a paradigm for reasoning under uncertainty,is then introduced as a general framework suitable for spatial analysis and decision making in a complex environment. In addition to the symbolic approaches, massively parallel distributed systems such as neural networks and genetic algorithms are proposed as powerful knowledge representation and inference models, especially for automatic acguisition of spatial knowledge through learning by examples. Lastly, the paper explores ways of using integratively and inter-actively spatial knowledge, unstructured and structured (such as mathematical and statistical models) ,and spatial information systems in spatial analysis and decision making. It also propos-es ways by which intelligent spatial decision support systems can be effectively built and applied to solve real-life decision problem.
【CateGory Index】： TP18