Granular Computing: Analysis and Design of Intelligent System

Book Preface

Undoubtedly, intelligent systems come with various connotations, exhibit different semantics, and imply numerous realizations. In spite of the existing diversity of views and opinions, there are several features that are quite profoundly visible across a broad spectrum of those systems. We anticipate that to some extent an intelligent system should be able to effectively communicate with the user, seamlessly accept requests from the human, and convey the results in a transparent, meaningful, and easy to comprehend format. A user-friendly, two-way communication between the human and the system arises as an imperative feature of intelligent systems. Furthermore, an intelligent system has to be endowed with substantial learning capabilities where learning can be carried out not only in the presence of numeric data but also in situations when the system is exposed to nonnumeric evidence, perceptions, opinions, and judgments. These forms of interaction and learning/adaptation are difficult to realize at the level of plain numbers but have to be carried out at the level of more abstract, carefully structured entitie —information granules. There are piles of data everywhere—in engineering, marketing, banking, Internet, insurance, chemical plants, and transportation networks. We face a genuine challenge: how to make sense of them, to help the user understand what they tell, and to gather useful evidence to support processes of decision making. It is apparent that these processes have to lead to constructs that form a useful abstraction of original data. In response to these growing challenges, Granular Computing (GrC) emerged a decade ago as a unified conceptual and processing framework.