Classifier Model Based on Neighbourhood Rough Set and Genetic Neural Network
COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12C) 853-858
1 School of Science, Guangdong University of petrochemical technology, Maoming Guangdong, China, 525000
In this study, the backpropagation (BP) neural network data classifier model optimized by genetic algorithm (GA) and neighborhood rough set is proposed. First, we integrate the attribute reduction technique with the neighborhood rough set, which is used to delete the redundant attributes of training samples. Second, we optimize the weights and thresholds of the BP neural network by using GA. As such, the training speed and generalization capability of the BP neural network are improved to obtain the optimal weights and thresholds. Finally, the experimental results show that the proposed algorithm performs well.