Modeling Urban Sprawl and Land Use Change in a Coastal Area-- A Neural Network Approach

Complexity of urban systems necessitates the consideration of interdependency among various factors for land use change modeling and prediction. The objective of this study is to explore the applicability of computational neural networks in modeling urban sprawl and land use change coupled with geographic information systems (GIS) in Hilton Head Island, South Carolina. We are particularly interested in the capabilities of neural networks to identify land use patterns, to model new development, and to predict future change. A binary logistic regression model is estimated comparison. The results indicate the neural network model is an improvement over the logistic regression model in terms of prediction accuracy.