The metallogeny of Central Iran is mainly characterized by the presence of several iron, apatite, and uranium deposits of Proterozoic age. Neural network method is used as a data-driven method for GIS-based predictive mapping of Proterozoic iron oxide (CU-U-AU-REE) mineralization in Central Iran. The radial based function link network (RBFLN) which is a modification of radial basis function neural networks (RBFNN) is employed as a neural network system. The evidential maps comprising of stratigraphic, structural, geophysical, and geochemical maps are used as n-dimensional vectors input to the RBFLN. A number of 58 deposits and 58 non-deposits are employed to train the network. The operations for the application of neural networks applied in this study involve both multiclass and binary representation of evidential maps. Running RBFLN on different input data shows that the increase in the number of evidential maps and classes leads to higher classification sum of squared error (SSE). As a whole the increase in the number of iterations results in the improvement of training SSE. The results of applying RBFLN show that a successful classification depends on the existence of well distributed deposit and non-deposit sites through the study area.