Rule extraction from trained neural networks using genetic algorithms
Proceedings of the second world congress on Nonlinear analysts: part 3
Template-based procedures for neural network interpretation
Neural Networks
A search technique for rule extraction from trained neural networks
Non-Linear Analysis
Extracting Provably Correct Rules from Artificial Neural Networks
Extracting Provably Correct Rules from Artificial Neural Networks
| Hi-index | 0.00 |
In this paper an evolutionary approach to the crisp rule extraction from a trained neural network for classification problems is described. The presented method is based on simultaneously working evolutionary algorithms. Each of them searches for rules from one class. After each generation the best rules are candidates to an updating a final set of rules describing behaviour of the trained neural network. The form of a chromosome and fitness function of the evolutionary algorithm is described. The results of experiments performed on benchmark data sets are discussed, as well.