Using Artifical Neural Networks in Landscape Architecture

Abstract

An artificial neural network (ANN) is a flexible mathematical structure based on the function of the human brain which is capable of identifying complex nonlinear relationships between input and output data sets. Artificial neural networks models have been found useful and efficient, particularly in problems for which the characteristics of the processes are difficult to describe using physical equations. They have been shown to be universal and highly flexible function approximators for any data. These are defined as powerful tools for models and an attractive alternative to the statistical classifiers, especially when the underlying data relationships are unknown. It is known that artificial neural network which is integrated with several softwares, have been used in landscape planning and land cover change detection with remote sensing applications for many years. With the development of new independent artificial neural network softwares, it is possible to create new decisionmaking approaches with the ability to implicitly detect complex nonlinear relationships between dependent and independent variables. Also with the ability to detect all possible interactions between predictor variables gives great opportunities in not only for landscape planning but also landscape design studies. In this study, an overview of the features of artificial neural networks and logistic regression is presented, opportunities and risks of artificial neural networks and the advantages and disadvantages of using this modeling technique in landscape architecture studies are discussed.