A Vectorization-Optimization-Method-Based Type-2 Fuzzy Neural Network for Noisy Data Classification

This paper proposes a vectorization-optimization-method (VOM)-based type-2 fuzzy neural network (VOM2FNN) for noisy data classification. In handling problems with uncertainties, such as noisy data, type-2 fuzzy systems usually outperform their type-1 counterparts. Hence, type-2 fuzzy sets are adopted in the antecedent parts to model the uncertainty. To consider the classification problems, the discriminative capability is crucial to determine the performance. Therefore, a VOM is proposed in the consequent parts to increase the discriminability and reduce the parameters. Compared with other existing fuzzy neural networks, the novelty of the proposed VOM2FNN is its consideration of both uncertainty and discriminability. The effectiveness of the proposed VOM2FNN is demonstrated by three classification problems. Experimental results and theoretical analysis indicate that the proposed VOM2FNN performs better than the other fuzzy neural networks.