ZHOU J. (1) ; CIVCO D. L. (1) ;
Photogrammetric engineering and remote sensing ISSN 0099-1112 CODEN PERSDV
1996, vol. 62
Traditional approaches for suitability analysis in GIS are overlay and the more complicated multicriteria evaluation (MCE). Despite being widely used, these methods have at least three problems: (1) difficulties in handling spatial data possessing inaccuracy, multiple measurement scales, and factor interdependency; (2) requirements of prior knowledge in identifying criteria, assigning scores, determining criteria preference, and selecting aggregation functions; and (3) typically, an unfriendly user interface. To solve these problems, in this paper a neural network approach is presented. The neural network uses a genetic algorithm as its learning mechanism. A set of experiments revealed that the aforementioned difficulties are overcome by the evolutionary learning of neural networks. Our conclusion is that genetic learning neural networks can provide an alternative for and improvement over traditional suitability analysis methods in GIS.