GIS and automatic learning tools for land use change analysis: an ANN application in multifunctional areas

Abstract

The paper aims to define a territorial analysis model for land use changes in multifunctional areas (MF) located near Florence in Tuscany region. Starting from existing project, Spatial Multicriteria Decision Analysis has been used with the aim of multifunctional degree definition of case study areas. Later land use changes have been performed by Artificial Neural Network model: using multitemporal analysis of land use, a non linear model Multi Layer Perceptron (MLP) neural network has been used in order to generate transition rules of land use and potential transition maps. Final step has generated a map of vulnerability to changes in MF areas.