Design of a Novel Neural Networks Based On Rough Sets

This paper proposed a novel neural network based on rough sets theory. Firstly a rule extraction method is discussed, and a set of rough rules is found from the given training data based on rough sets theory. Secondly, the structure and model of the neural network are designed according to these rules, and the training algorithm with high-precision of learning is formulated based on neural networks techniques. Compared with the conventional neural network, the proposed neural network has the following advantages: good understandability, simple computation and high-precision. Finally, a lot of numerical simulations have been conducted, and simulation results have shown that the novel neural network is feasible and efficient in function approximation. The novel neural network has great potential in the application areas of signal processing, pattern recognition, process modeling and implementation of high-precision real-time intelligent controller.