Today's networked world and the decentralization that the Web enables and symbolizes have created new phenomena: information explosion and saturation. To deal with information overload, our computers should have human-centered functionality and enhanced intelligence, but instead they simply become faster. Soft computing is a unifying framework that combines techniques in neural networks, fuzzy theory, genetic algorithms, and artificial intelligence to develop intelligent systems able to learn in dynamic, imprecise, and uncertain environments. This book explains the theory, methodology, and application aspects of human-centered systems, showing how it is possible to extend to machines such techniques as dynamic cognitive learning, neural-fuzzy-based learning, and genetic-evolutionary type learning paradigms.