Estimating Garson weights from neural networks

Based on Gevrey et al. (2003), this function estimates the weights of the nodes/loci in the neural networks. For each hidden neuron i, multiply the absolute value of the hidden-output layer connection weight by the absolute value of the hidden-input layer connection weight. Do this for each input variable j. The following products Pij are obtained.

GarsonWeights(object)

Arguments

object	The neural netwrok model from one of the h2o, RFCNN4, RSNNS models.

Details

Based on Gevrey et al. (2003), this function estimates the weights of the nodes in the neural networks. For each hidden neuron i, multiply the absolute value of the hidden-output layer connection weight by the absolute value of the hidden-input layer connection weight. Do this for each input variable j. The following products Pij are obtained.

Value

The nueral weights.

References

Gevrey, M., Dimopoulos, I., & Lek, S. (2003). Review and comparison of methods to study the contribution of variables in artificial neural network models. Ecological modelling, 160(3), 249-264.