作者: Farshid Farnood Ahmadi , Nasir Farsad Layegh
DOI: 10.1007/S00521-014-1801-Z
关键词:
摘要: The main objective of this study is to investigate the potential artificial neural network (ANN) for integration with geographical information system (GIS) assess suitability land cultivate a selected crop. For purpose, requirements intelligent assessment suitably were determined and architecture integrated was designed according capabilities ANN GIS. evaluation performance, case carried out in Mazandaran Province, located northern part Iran. In implemented study, four effective parameters crop cultivation utilized multilayer feed-forward back-propagation analysis. activation function hidden layer set log-sigmoid function. Designed trained using samples extracted from GIS database, 4-6-1 (4 input as layer, 6 neurons 1 target value representing level) structure shows best result. Evaluation that 83.43 % ANN's results are acceptable consistent real world.