Input space configuration effects in neural network-based grade estimation

Authors:
I. K. Kapageridis
Affiliations:
Department of Geotechnology and Environmental Engineering, School of Technological Applications, Technological Education Institute of West Macedonia, Koila, Kozani GR 501 00, Greece
Venue:
Computers & Geosciences
Year:
2005

Citing 5
Cited 2

Radial basis functions for multivariable interpolation: a review

Algorithms for approximation
Reserve estimation using neural network techniques

Computers & Geosciences
Approximation and radial-basis-function networks

Neural Computation
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Fast learning in networks of locally-tuned processing units

Neural Computation

LavaNet-Neural network development environment in a general mine planning package

Computers & Geosciences
The application of median indicator kriging and neural network in modeling mixed population in an iron ore deposit

Computers & Geosciences

Quantified Score

Hi-index	0.00

Visualization

Abstract

The way input data are presented to Artificial Neural Networks is one of the most important parameters controlling their performance during the development and application stages. The choice of dimensions that form the input space of a network (dimensionality) is very important and must be investigated as to its effects on the performance of artificial neural network systems applied to grade estimation. The study of these effects was achieved by configuring the available data in order to form different multidimensional input spaces and testing on different datasets. The results obtained from the numerous tests in this study lead to a better understanding of the behaviour of artificial neural network systems when facing different input space configurations using the same data, and aid the choice of dimensions that will allow better representation of samples for their development for grade estimation.