A novel data-mining method based on the IMVFEWNN

Based on analyzing the immune phenomena in nature and utilizing the performance of the existent artificial neural network, a novel network structure, i.e. IMVFEWNN is proposed which integrates the immune mechanism and the structure of neural information processing. The learning algorithm of MVFEWNN is also given which contains the method of selecting an activation function and the adaptive algorithm of the network. This model makes it easy for a user to directly utilize the characteristic information of a pending problem and to simplify the original structure through adjusting the activation function with the prior knowledge, and then the working efficiency and the searching accuracy are both improved. The analysis in theory and the simulating test for the data mining problem show that, compared with the artificial neural network, IMVFEWNN is not only effective but also feasible.