JOURNAL OF INTELLIGENT SYSTEMS

 

Vol. 14: No. 2/3, 2005

SPECIAL ISSUE

 

hybrid intelligent systems for time series prediction

Guest Editors: Oscar Castillo and Patricia Melin

 

ã 2005, Published by:

Freund & Pettman Publishers

Tel Aviv/London

Price: $120, including air mail

 

contents

000000

Preface

O. Castillo and P. Melin

 

 

Stepwise Selection of Artificial Neural Network Models for  Time Series Prediction

S.F. Crone

 

 

Grammar-Mediated Time-Series Prediction

A. Brabazon, K. Meagher, E. Carty, M. O’Neill, P. Keenan

 

 

Predictive Neural Model of an Osmotic Dehydration Process

I. Baruch, P. Genina-Soto and J. Barrera-Cortés

 

 

Evolutionary Multiresolution Filtering to Forecast Nonlinear Time Series

E. Gómez-Ramírez & A. Ayala-Hernández

 

 

Simulation and Forecasting Complex Economic Time Series Using Neural Network Models

P. Melin, O. Castillo, A. Mancilla and M. Lopez

 

 

Human Evolutionary Model

O. Montiel, O. Castillo, P. Melin, A. Rodriguez-Diaz and R. Sepulveda

 

 

Handling Uncertainty in Controllers Using Type-2 Fuzzy Logic

R. Sepulveda, O. Castillo, P. Melin, O. Montiel and A. Rodriguez-Diaz

 

 

 

Soft computing (SC) can be used to build hybrid intelligent systems for achieving different goals in real-world applications. Soft-computing techniques include fuzzy logic, neural networks, genetic algorithms, chaos theory methods, and similar techniques. Each approach has advantages and disadvantages, and several real-world problems have been solved by using one of these techniques. Nevertheless, many real-world complex problems require the integration of several techniques to achieve the efficiency and accuracy needed in practice. Genetic algorithms can be used to optimize the topology of a fuzzy or a neural system. Neuro-fuzzy approaches or even neuro-fuzzy-genetic approaches exist for designing the best intelligent system for a particular application. This Special Issue consists of seven papers that consider the use and integration of different SC techniques for developing hybrid intelligent systems for time series prediction. Stepwise Selection of Artificial Neural Network Models for Time Series Prediction by Crone presents a new approach that enables the selection of appropriate neural network models for specific applications. Grammar-Mediated Time Series Prediction by Brabazon and coworkers, deals with a new approach using grammatical evolution for achieving time series prediction to uncover useful technical trading rule-sets for intra-day equity trading. Predictive Neural Model of Osmotic Dehydration Process by Baruch and colleagues, describes the use of a recurrent neural network model for predicting the osmotic dehydration kinetics of nature product cubes. Evolutionary Multiresolution Filtering to Forecast Nonlinear Time Series by Gomez-Ramirez and Ayala-Hernandez proposes a new adaptive scheme of multiresolution filtering to decompose a time series into simpler ones. The scheme proposed in this paper uses a genetic algorithm to find the optimal bank of filters without previous knowledge of the behavior of the system to be identified. Simulation and Forecasting Complex Economic Time Series Using Neural Networks by Melin and coworkers describes a new approach for simulating and forecasting time series using different neural network models. Simulation results show that modular neural networks clearly outperform monolithic ones, both in accuracy and efficiency. Human Evolutionary Model by Montiel and colleagues proposes a new computational model that enhances evolutionary models with human characteristics. This model uses intelligence and intuition as the main driving forces. Handling Uncertainty in Controllers Using Type-2 Fuzzy Logic by Roberto Sepulveda and coworkers, deals with the application of type-2 fuzzy logic for achieving intelligent control of non-linear dynamical systems. In our opinion, all the papers in this special issue make an important contribution to the state of the art in the field of hybrid intelligent systems, in general, and to the areas of fuzzy logic, neural networks, and genetic algorithms, in particular. We sincerely hope that this issue will be of great interest and use to researchers and students all over the world.

 

Oscar Castillo and Patricia Melin

 

ORDER FORM

Name: ..........................................................................................Department:.................................................................................

Address: ...........................................................................................................................................................................................

Signature.......................................................................................Telephone:.................................................................................

Fax:...............................................................................................E-mail:...........................................................................................

 

Please return to: Freund Publishing House Ltd.
P.O. Box 35010, Tel Aviv 61350, Israel; e-mail: h_freund@netvision.net.il ; www.freundpublishing.com

 

 

 

 

 

 

 

I would like to order _____ copies of Journal of Intelligent Systems, Special Issue on Consciousness

 

 

 

Vol. 14: 2/3, 2005, at a cost of $120 (including air mail)

 

 

 

I enclose a check (made payable to Freund Publishing House Ltd.) for the amount of ________________

 

 

 

Please debit my VISA/Mastercard/American Express card. no.....................................Expiry date..............