Applications of Bayesian Classifiers: Uncertainty Modeling, Global Optimization, and Solidmechanics Problems

In this book, it is shown how pattern recognitionapproaches developed in computer science can beseamlessly combined with statistical techniques toproduce a knowledge driven search algorithm. Twodifferent classes of applications are used todemonstrate the wide applicability of the algorithm. The first class of applications is in global designoptimization. In design optimization applications, Bayesian classification tree is used to generatehigh-performing design alternatives. The second is inthe structural mechanics of systems with spatiallyrandom properties. In applications involvingspatially random properties, the classification treeprovides pattern-based features that can be used toidentify regions of a system with desirable orundesirable local properties. The theories andexample applications should be especially useful toresearchers and professionals in the field ofuncertainty modeling and design optimization.