Neural Network Ensembles Based Approach for Mineral Prospectivity Prediction

In mining industry, accurate identification of new geographic locations that are favourable for mineral exploration is very important. However, definitive prediction of such locations is not an easy task. In recent years, the use of neural networks ensemble approach to the classification problem has gained much attention. This paper discusses the results obtained from using different neural network (NN) ensemble techniques for the mineral prospectivtity prediction problem. The proposed model uses the geographic information systems (GIS) data of the location. The method is tested on the GIS data for the Kalgoorlie region of Western Australia. The results obtained are compared to some of the commonly known techniques: the majority combination rule, averaging technique, weighted averaging method tuned by genetic algorithm (GA) and a newly proposed rule based method. The results obtained using the different techniques are discussed.