Case-based cellular automaton for simulating urban development in a large complex region

The essential part of geographical cellular automata (CA) is to provide appropriate transition rules so that realistic patterns can be simulated. Transition rules can be defined by a variety of methods, such as multicriteria evaluation (MCE), logistic regression, neural networks, and data mining. The solicitation of concrete knowledge (transition rules) is often difficult for many applications. There are problems in representing complex relationships by using detailed rules. This study demonstrates that the case-based approach can avoid the problems of the rule-based approach in defining CA. The proposed method is based on the case-based reasoning techniques, which don't require the procedure of soliciting explicit transition rules. The knowledge for determining the state conversion of CA is inexplicitly embedded in discrete cases. The lazy-learning technology can be used to represent complex relationships more effectively than detailed equations or explicit transition rules. This paper presents an extended cellular automaton in which transition rules are represented by using case-based reasoning (CBR) techniques. The common k-NN algorithm of CBR has been modified to incorporate the location factor to reflect the spatial variation of transition rules. Multi-temporal remote sensing images are used to obtain the adaptation knowledge in the temporal dimension. This model has been applied to the simulation of urban development in the Pearl River Delta which has a hierarchy of cities. Comparison indicates that this model can produce more plausible results than rule-based CA in simulating large complex regions.