Inspired by [4] and in the quest to remedy the
limitations associated with the scheme, this paper
focuses on further tuning the ANN model. Specifically,
based on one of the most extensive and recent
GIS data available, the Multi Layer Perceptron
(MLP) and the Radial Basis Function (RBF) structures
were experimented. The underlying rationale
regarding each constituent factor will be discussed.
In addition, guidelines and insight precautions on applying
the ANN on predictions of flood and probably
similar disasters were also addressed.