# Fuzzy Classifier Design

## Kuncheva, L.I.

#### University of Wales, Bangor, Gwynedd, UK

ISBN : 3-7908-1298-6

(Series: Studies in Fuzziness and Soft Computing, Vol. 49)

This book about fuzzy classifier design briefly introduces the fundamentals of supervised pattern recognition and fuzzy set theory. Fuzzy if-then classifiers are defined and some theoretical properties thereof are studied. Popular training algorithms are detailed. Non if-then fuzzy classifiers include relational, k-nearest neighbor, prototype-based designs, etc. A chapter on multiple classifier combination discusses fuzzy and non-fuzzy models for fusion and selection.

Keywords: Fuzzy Logic, Fuzzy Classifier Design, Pattern Recognition

Contents:
Introduction: What are fuzzy classifiers?- The data sets used in this book.- Notations and acronyms.- Organization of the book.- Acknowledgements.-
Statistical Pattern Recognition: Class, feature, feature space. - Classifier, discriminant functions, classification regions.- Clustering.- Prior probabilities, class-conditional probability density functions, posterior probabilities.- Minimum error and minimum risk classification. Loss matrix.- Performance estimation.- Experimental comparison of classifiers.- A taxonomy of classifier design methods.-
Statistical Classifiers: Parametric classifiers.- Nonparametric classifiers.- Finding k-nn prototypes.- Neural networks.-
Fuzzy Sets: Fuzzy logic, an oxymoron?- Basic definitions.- Operations on fuzzy sets.- Determining membership functions.-
Fuzzy If-then Classifiers: Fuzzy if-then systems.- Function approximation with fuzzy if-then systems.- Fuzzy if-then classifiers.- Universal approximation and equivalences of fuzzy if-then classifiers.-
Training of Fuzzy If-then Classifiers: Expert opinion or data analysis.- Tuning the consequents.- Tuning the antecedents.- Tuning antecedents and consequents using clustering.- Genetic algorithms for tuning fuzzy if-then classifiers.- Fuzzy classifiers and neural networks: hybridization or identity?- Forget interpretability and choose a model.-
Non if-then Fuzzy Models: Early ideas.- Fuzzy k-nearest neighbors (k-nn) designs.- Generalized nearest prototype classifier (GNPC).-
Combinations of Multiple Classifiers Using Fuzzy Sets: Combining classifiers: the variety of paradigms.- Classifier Selection.- Classifier Fusion.- Experimental results.-
Conclusions: What to Choose