Qiang Shen – författare

Computational Intelligence and Feature Selection provides readers with the background and fundamental ideas behind Feature Selection (FS), with an emphasis on techniques based on rough and fuzzy sets. For readers who are less familiar with the subject, the book begins with an introduction to fuzzy set theory and fuzzy-rough set theory. Building on this foundation, the book provides:

A critical review of FS methods, with particular emphasis on their current limitations

Program files implementing major algorithms, together with the necessary instructions and datasets, available on a related Web site

Coverage of the background and fundamental ideas behind FS

A systematic presentation of the leading methods reviewed in a consistent algorithmic framework

Real-world applications with worked examples that illustrate the power and efficacy of the FS approaches covered

An investigation of the associated areas of FS, including rule induction and clustering methods using hybridizations of fuzzy and rough set theories

Computational Intelligence and Feature Selection is an ideal resource for advanced undergraduates, postgraduates, researchers, and professional engineers. However, its straightforward presentation of the underlying concepts makes the book meaningful to specialists and nonspecialists alike.

The book is a timely report on advanced methods and applications of computational intelligence systems. It covers a long list of interconnected research areas, such as fuzzy systems, neural networks, evolutionary computation, evolving systems and machine learning. The individual chapters are based on peer-reviewed contributions presented at the 16th Annual UK Workshop on Computational Intelligence, held on September 7-9, 2016, in Lancaster, UK. The book puts a special emphasis on novels methods and reports on their use in a wide range of applications areas, thus providing both academics and professionals with a comprehensive and timely overview of new trends in computational intelligence.

This book chiefly presents a novel approach referred to as backward fuzzy rule interpolation and extrapolation (BFRI). BFRI allows observations that directly relate to the conclusion to be inferred or interpolated from other antecedents and conclusions. Based on the scale and move transformation interpolation, this approach supports both interpolation and extrapolation, which involve multiple hierarchical intertwined fuzzy rules, each with multiple antecedents. As such, it offers a means of broadening the applications of fuzzy rule interpolation and fuzzy inference. The book deals with the general situation, in which there may be more than one antecedent value missing for a given problem. Two techniques, termed the parametric approach and feedback approach, are proposed in an attempt to perform backward interpolation with multiple missing antecedent values. In addition, to further enhance the versatility and potential of BFRI, the backward fuzzy interpolation method isextended to support α-cut based interpolation by employing a fuzzy interpolation mechanism for multi-dimensional input spaces (IMUL). Finally, from an integrated application analysis perspective, experimental studies based upon a real-world scenario of terrorism risk assessment are provided in order to demonstrate the potential and efficacy of the hierarchical fuzzy rule interpolation methodology. 

This book covers a comprehensive approach to the development and application of a suite of novel algorithms for practical approximate knowledge-based inference. It includes an introduction to the fundamental concepts of fuzzy sets, fuzzy logic, and fuzzy inference. Collectively, this book provides a systematic tutorial and self-contained reference to recent advances in the field of fuzzy rule-based inference. 

Approximate reasoning systems facilitate inference by utilizing fuzzy if-then production rules for decision-making under circumstances where knowledge is imprecisely characterized. Compositional rule of inference (CRI) and fuzzy rule interpolation (FRI) are two typical techniques used to implement such systems. The question of when to apply these potentially powerful reasoning techniques via automated computation procedures is often addressed by checking whether certain rules can match given observations. Both techniques have been widely investigated to enhance the performance of approximate reasoning. Increasingly more attention has been paid to the development of systems where rule antecedent attributes are associated with measures of their relative significance or weights. However, they are mostly implemented in isolation within their respective areas, making it difficult to achieve accurate reasoning when both techniques are required simultaneously.

This book first addresses the issue of assigning equal significance to all antecedent attributes in the rules when deriving the consequents. It presents a suite of weighted algorithms for both CRI and FRI fuzzy inference mechanisms. This includes an innovative reverse engineering process that can derive attribute weightings from given rules, increasing the automation level of the resulting systems. An integrated fuzzy reasoning approach is then developed from these two sets of weighted improvements, showcasing more effective and efficient techniques for approximate reasoning. Additionally, the book provides an overarching application to interpretable medical risk analysis, thanks to the semantics-rich fuzzy rules with attribute values represented in linguistic terms. Moreover, it illustrates successful solutions to benchmark problems in the relevant literature, demonstrating the practicality of the systematic approach to weighted approximate reasoning.