Computational Intelligence Systems and Applications

1 Introduction.- 1.1 A general concept of computational intelligence.- 1.2 The building blocks of computational intelligence systems.- 1.3 Objectives and scope of this book.- 2 Elements of the theory of fuzzy sets.- 2.1 Basic notions, operations on fuzzy sets, and fuzzy relations.- 2.2 Fuzzy inference systems.- 3 Essentials of artificial neural networks.- 3.1 Processing elements and multilayer perceptrons.- 3.2 Radial basis function networks.- 4 Brief introduction to genetic algorithms.- 4.1 Basic components of genetic algorithms.- 4.2 Theoretical introduction to genetic computing.- 5 Main directions of combining artificial neural networks, fuzzy sets and evolutionary computations in designing computational intelligence systems.- 5.1 Artificial intelligence versus computational intelligence.- 5.2 Designing computational intelligence systems.- 5.3 Selected neuro-fuzzy systems.- 5.3.1 ANFIS system.- 5.3.2 NEFCLASS system.- 5.3.3 NEFPROX system.- 5.3.4 Neuro-fuzzy system of [242].- 6 Neuro-fuzzy(-genetic) system for synthesizing rule-based knowledge from data.- 6.1 Synthesizing rule-based knowledge from data — statement of the problem.- 6.2 Neuro-fuzzy system in learning mode — problem of knowledge acquisition.- 6.2.1 Conceptual scheme of the system.- 6.2.2 Implementation of the system.- 6.3 Neuro-fuzzy system in inference mode — approximate inference engine.- 6.3.1 Concept of the system.- 6.3.2 Implementation of the system.- 6.3.3 Testing and pruning the system.- 6.4 Learning techniques.- 6.4.1 Backpropagation-like method.- 6.4.2 Optimization techniques.- 6.4.2.1 Conjugate-gradient algorithm.- 6.4.2.2 Variable-metric algorithm.- 6.4.3 Genetic algorithms.- 6.5 A numerical example of synthesizing rule-based knowledge from data — modelling the Mackey-Glass chaotic time series.- 6.5.1 Designing the neuro-fuzzy model from data.- 6.5.2 A comparative analysis with several alternative modelling techniques.- 6.6 Synthesizing rule-based knowledge from “fish data”.- 6.6.1 Designing the neuro-fuzzy-genetic system from data.- 6.6.2 A comparison with other methodologies.- 7 Rule-based neuro-fuzzy modelling of dynamic systems and designing of controllers.- 7.1 System identification — statement of the problem and its general solution in the framework of neuro-fuzzy methodology.- 7.2 Rule-based neuro-fuzzy modelling of an industrial gas furnace system.- 7.2.1 Designing the neuro-fuzzy model of dynamic system from data.- 7.2.2 A comparative analysis with several alternative methodologies.- 7.3 Designing the neuro-fuzzy controller for a simulated backing up of a truck.- 7.3.1 Designing the controller from data.- 7.3.2 A comparison of different neuro-fuzzy controllers.- 8 Neuro-fuzzy(-genetic) rule-based classifier designed from data for intelligent decision support.- 8.1 Designing the classifier from data — statement of the problem.- 8.2 Learning mode of neuro-fuzzy classifier.- 8.2.1 Conceptual scheme of the classifier.- 8.2.2 Implementation of the classifier.- 8.3 Inference (decision making) mode of neuro-fuzzy classifier.- 8.3.1 Concept of the system and its implementation.- 8.3.2 Testing and pruning the system.- 8.4 Neuro-fuzzy decision support system for diagnosing breast cancer.- 8.4.1 Designing the system from data.- 8.4.2 A comparative analysis of several different methodologies applied to diagnosing breast cancer.- 8.5 Neuro-fuzzy-genetic decision support system for the glass identification problem (forensic science).- 8.5.1 Designing the system from data.- 8.5.2 A comparative analysis with other techniques for decision support systems design.- 8.6 Neuro-fuzzy-genetic decision support system for determining the age of abalone (marine biology).- 8.6.1 Designing the system from data.- 8.6.2 A comparative analysis with alternative approaches.- 9 Fuzzy neural network for system modelling and control.- 9.1 Learning mode of the network.- 9.2 Inference mode of the network.- 9.3 Fuzzy neural modelling of dynamic systems (an industrial gas furnace system).- 9.4 Fuzzy neural controller.- 9.4.1 Structure, learning and operation of the controller.- 9.4.2 A numerical example of fuzzy neural control.- 10 Fuzzy neural classifier.- 10.1 Learning and inference modes of the classifier.- 10.2 Fuzzy neural classifier for diagnosis of surgical cases in the domain of equine colic.- A Appendices.- A.1.1 Inputs.- A.1.2 Output.- A.2.1 Inputs.- A.2.2 Outputs — set of two class labels.- A.3.1 Inputs.- A.3.2 Outputs — set of two class labels.- A.4.1 Inputs.- A.4.2 Outputs — set of three class labels.- A.5.1 Inputs.- A.5.2 Outputs — three sets of class labels.- References.