Neural network, genetic, and fuzzy logic models of spatial interaction

The author investigates the extent to which smart computational methods can be used to create new and better performing types of spatial interaction model. He briefly describes the application of three different computationally intensive modelling technologies and compares the performance of the resulting models on a benchmark data set. It would appear that performance improvements of up to a factor of two can be obtained at the cost of a few orders of magnitude increase in compute times.