Neural Networks In A Softcomputing Framework

Neural Networks in a Softcomputing Framework

Conventional model-based data processing methods are computationally expensive and require experts’ knowledge for the modelling of a system. Neural networks are a model-free, adaptive, parallel-processing solution. This textbook provides a powerful and universal paradigm for information processing; it reviews the most popular neural-network methods and their associated techniques. Each chapter has a systematic survey of each neural-network model. Computational intelligence topics like fuzzy logic and genetic algorithms (tools for neural-network learning) are introduced. Array signal processing problems are used to show the applications of each model. This is an ideal textbook for graduate students and researchers; as well as introducing the basics, the exhaustive list of references included will aid their future research. It is also a valuable reference for scientists and practitioners working in pattern recognition, signal processing, speech and image processing, data analysis and A.I.

Neural Networks In A Softcomputing Framework

Recurrent Neural Networks and Soft Computing

New applications in recurrent neural networks are covered by this book, which will be required reading in the field. Methodological tools covered include ranking indices for fuzzy numbers, a neuro-fuzzy digital filter and mapping graphs of parallel programmes. The scope of the techniques profiled in real-world applications is evident from chapters on the recognition of severe weather patterns, adult and foetal ECGs in healthcare and the prediction of temperature time-series signals. Additional topics in this vein are the application of AI techniques to electromagnetic interference problems, bioprocess identification and I-term control and the use of BRNN-SVM to improve protein-domain prediction accuracy. Recurrent neural networks can also be used in virtual reality and nonlinear dynamical systems, as shown by two chapters.