Pseudo outer-product based fuzzy neural networks

Abstract

Fuzzy rule-based systems and neural networks are complementary Artificial Intelligence (AI) techniques. Considerable amount of research and development has been directed at the integration of fuzzy rule-based systems and neural network techniques into hybrid fuzzy neural networks. Such hybrid systems enhance the cumulative advantages and alleviate the shortcomings of the respective techniques. Several novel fuzzy neural network architectures and their learning algorithms have been developed as part of this doctoral work and are presented in this thesis.