Over the last two decades, there have been many substantive developments in the field of artificial intelligence (AI). One rapidly evolving aspect of this field involves methods to address new problems without relying on existing human expertise. These methods conform to a broad set of principles gleaned from learning systems observed in nature, including neural networks, fuzzy systems, evolutionary computing, and related techniques. These collective techniques form the basis for a new field, loosely known as computational intelligence. A subset of this field, using methods that primarily model human information processing, is called soft computing. One important goal of this field is to develop fundamental understanding and generalized tool sets, using soft computing and computational intelligence methodologies, to make possible intelligent self-learning systems of increasing complexity and skill levels that may in fact create something new.
This book presents a collection of recent papers that survey the state of the art in this field, fusing soft computing with traditional computational methodologies (known as hard computing) to provide hybrid methods that are found to be more effective than either methodology alone. The editor has assembled 11 papers for this text, combining the experience of a wide range of soft computing experts, to illustrate applications of various types of hybrid methods across diverse problem sets. While not an encyclopedic compendium, these well-written papers serve to offer insight into the powerful combination of soft and hard computing that is now being applied, with increasing capability, to real-world applications.
Each paper is presented as a standalone chapter, with each having a short introduction by the editor to add an overview and context. Within each chapter, these introductory remarks, and the papers themselves, have separate lists of references, to assist the reader in further pursuing specific details. The first paper, providing a background for, and a developmental overview of, this rapidly emerging field, is authored by the editor himself. The ten subsequent papers serve to illustrate the wide-ranging applicability of methodological fusion and integration schema in selected multi-disciplinary application environments. These papers are generally application specific, tending to be somewhat less theoretical in nature, and provide numerous examples of practical implementation issues and applied research results. The application domains discussed in these ten chapters cover many diverse disciplines, including aerospace, computer security, data mining, robust control, tool we!
ar monitoring, and control of electric power systems and large-scale industrial plants.
These 11 individual contributions seem to be well selected for the intended purpose of this book. Given that the fusion of soft and hard computing, and the resulting computationally intelligent hybrid systems are essentially linked to applications, the editor’s introduction to each paper was especially helpful in understanding the importance and cross-disciplinary applicability of each contribution. Overall, this collection of well-edited papers seems most appropriate for use by practicing engineers, researchers, research and development managers, and graduate students in these fields who want to deepen their knowledge about this emergent and increasingly important topic.