Soft Computing Based Modeling in Intelligent Systems (Studies in Computational Intelligence, 196)

The book includes soft computing implementations of intelligent systems models. The recent popularity of fuzzy systems, neural networks and evolutionary computation, considered as related in AI, are now widely used to build intelligent systems. Professor Lotfi A. Zadeh has suggested the term "Soft Computing" for all new techniques working in these new areas of AI. Soft Computing techniques are tolerant to imprecision, uncertainty and partial truth. Due to the large variety and complexity of the domain, the constituting methods of Soft Computing are not competing for a comprehensive ultimate solution. Instead they are complementing each other, for dedicated solutions adapted to each specific problem. Hundreds of concrete applications are already available in many domains. Model based approaches offer a very challenging way to integrate a priori knowledge into procedures. Due to their flexibility, robustness, and easy interpretability, the soft computing applications will continue to have an exceptional role in our technologies. The book “Soft Computing Based Modeling in Intelligent Systems”contains the - tended works originally presented at the IEEE International Workshop SOFA 2005 and additional papers. SOFA, an acronym for SOFt computing and Applications, is an international wo- shop intended to advance the theory and applications of intelligent systems and soft computing. Lotfi Zadeh, the inventor of fuzzy logic, has suggested the term “Soft Computing.” He created the Berkeley Initiative of Soft Computing (BISC) to connect researchers working in these new areas of AI. Professor Zadeh participated actively in our wo- shop. Soft Computing techniques are tolerant to imprecision, uncertainty and partial truth. Due to the large variety and complexity of the domain, the constituting methods of Soft Computing are not competing for a comprehensive ultimate solution. Instead they are complementing each other, for dedicated solutions adapted to each specific pr- lem. Hundreds of concrete applications are already available in many domains. Model based approaches offer a very challenging way to integrate a priori knowledge into procedures. Due to their flexibility, robustness, and easy interpretability, the soft c- puting applications will continue to have an exceptional role in our technologies. The applications of Soft Computing techniques in emerging research areas show its mat- ity and usefulness. The IEEE International Workshop SOFA 2005 held Szeged-Hungary and Arad- Romania in 2005 has led to the publication of these two edited volumes. This volume contains Soft Computing methods and applications in modeling, optimisation and prediction. Front Matter....Pages - Soft Computing Methods for Global, Local and Personalized Modeling and Applications in Bioinformatics....Pages 1-18 Orthosymmetry Aspects of Aggregation Functions....Pages 19-34 MOGA Design of Neural Network Predictors of Inside Temperature in Public Buildings....Pages 35-61 Model Based Anytime Soft Computing Approaches in Engineering Applications....Pages 63-92 A Multi-structure Modeling Methodology....Pages 93-113 The Generalized Dombi Operator Family and the Multiplicative Utility Function....Pages 115-131 Multi-channel Representations of Max-Plus Algebra Based Wavelet Transform and Their Application to Video Coding....Pages 133-144 The Fuzzy-Interpolative Methodology....Pages 145-167 Interpolative-Type Control Solutions....Pages 169-203 Back Matter....Pages -