An immune neural network used for classification

Based on an analysis of immune phenomena in nature and utilizing performances of ANN, a novel network mode, i.e., an immune neural network (INN), is proposed which integrates the immune mechanism and the function of neural information processing. The learning algorithm of an INN selects an excitation function and adaptive algorithm of the network. This model makes it easy for a user to utilize directly the characteristic information of a problem and to simplify the original structure by adjusting the excitation function with prior knowledge, improving the working efficiency and searching accuracy. A theoretical analysis and a simulation test for the twin-spiral problem show that, compared with an artificial neural network, INN is not only effective but also feasible. INN can simplify the structure of the existent model and show good working performance