Cellular automata urban expansion model based on support vector machines

Land-use change models are used to explore the dynamics and drivers of land-use/landcover change and to inform policies affecting such change. A broad array of applications and modeling methods are available and each type has certain advantages and disadvantages depending on the objective of the research. This work presents an approach combining cellular automata (CA) model and supported vector machine (SVM) and binary logistic regression model (Logit) for simulating urban expansion in Wallonia (Belgium). This article emphasizes the interest in comparing combining CA with conventional Logit versus combining CA with SVM method as a base of CA model transition rule. Relative operating characteristic (ROC) and spatial matrices are used to validate the model. Model validation shows that the allocation performance of CA-SVM outperformed CALogit approach.