Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Computational intelligence : revised and selected papers of the International Joint Conference, IJCCI 2013, Vilamoura, Portugal, September 20-22, 2013
Madani K., Dourado A., Rosa A., Filipe J., Kacprzyk J., Springer International Publishing, New York, NY, 2015. 518 pp. Type: Book (978-3-319233-91-8)
Date Reviewed: Jul 26 2016

This book is composed of selected papers that were submitted and presented at the Fifth International Joint Conference on Computational Intelligence (IJCCI), held in Portugal in 2013. The conference is divided into three categories: evolutionary computation theory and applications (ECTA), fuzzy computation theory and applications (FCTA), and neural computation theory and applications (NCTA).

The first part of the book consists of ten papers. This review discusses three that I found very interesting. In “Incremental Hough Transform: A New Method for Circle Detection,” the authors present an extension to the circle Hough transform (CHT) algorithm, which they call incremental CHT (ICHT), where they include the use of trigonometric functions in order to recognize the shape of a circle. The CHT algorithm works by mapping the image space to a parameter space of curves that fit the feature points of the image. The use of trigonometric functions makes ICHT easy to use and fast in computational performance. The hardware implementation of CHT, the algorithm of ICHT, error analysis, and a comparison between CHT and ICHT processing times are well presented.

The second paper that I found interesting in this part is “Self-Adaptive Evolutionary Many-Objective Optimization Based on Relation &egr;-Preferred.” It is a single-author paper that provides much valuable content. The paper tackles the case of an optimization process that has multiple objectives (more than three). It is based on the &egr;-preferred method, which was also invented by the same author. &egr; is a parameter that distinguishes the best solutions. By using an adaptive evolutionary algorithm, the best values of &egr; can be found. Furthermore, the author considers the nurse rostering problem--the time scheduling of nurses’ shifts in a hospital depending on their availability and their calendar priorities--in order to examine the performance of his new algorithm. As already mentioned, rich and valuable content, depicted with mathematical theory and analysis, support the work. I recommend this paper to anyone interested in the problem of time scheduling that depends on user constraints and specifications.

“A Radial Basis Function Neural Network-Based Coevolutionary Algorithm for Short-Term to Long-Term Time Series Forecasting” is the third paper in the context of evolutionary computation that I will discuss. As mentioned in its title, the paper deals with time series prediction and forecasting. The major problem in predicting the evolution or the future values of a time sequence is finding the time lags (that is, time slots or time periods) that constitute and best describe the sequence being analyzed. Thus, the authors’ work is based on the L-Co-R algorithm, which they previously developed. L-Co-R is an evolutionary algorithm that is composed from two populations: the first population aims to find the best radial basis function neural network in terms of number of neuron layers, weights, and so on; the second population aims to find the best lags in order to predict future values. The authors use a variety of testing datasets for experimentation and compare the forecasting capability of their algorithm, in terms of efficiency and accuracy, with state-of-the-art algorithms in the literature. Their algorithm shows the best results.

The second set of 11 papers is concerned with fuzzy computation and applications. In this category, I will discuss “Gene Priorization for Tumor Classification Using an Embedded Method,” which I find very important since it deals with cancer diagnosis. The authors implement a fuzzy tree ensemble in order to classify a tumor based on gene expression and identification. The authors give an overview of the existing techniques to analyze gene expression data, like clustering analysis using self-organizing maps (SOM) or support vector machines (SVM), feature selection using the random forest method, and so on. Then, the authors discuss their fuzzy approach, which consists of building a fuzzy ensemble tree composed of three modules: the first is fuzzy random classification, which is responsible for learning and classification; the second ensemble is a filter and wrapper responsible for scaling and discretizing the feature points in the dataset; and the third ensemble is a technique for the final selection of an optimal subset of features of gene expressions. Colon cancer, leukemia, and prostate datasets are used in the study, and a detailed analysis of the results is provided.

The third part of the book consists of seven papers on neural computation. The authors of “Unsupervised Analysis of Morphological ECG Features for Attention Detection” make a link between EEG and ECG data when users perform attention-based tasks. First, they implement a filtering technique to perform feature extraction of both EEG and ECG input. Then, they apply unsupervised learning to perform clustering. Finally, cluster validation is implemented using a validation index in order to measure the goodness of a cluster. Experimental results are provided in order to analyze the relation between EEG and ECG data.

This book contains many interesting papers, the majority of which are up to the level of journal publications. Due to the limits set on review length, I was obliged to narrow down my selection. However, ideally I would present them all to you. If you are interested in computational intelligence, this book will give you a grasp of the latest advancements, interesting topics, and current challenges in the field.

Reviewer:  Mario Antoine Aoun Review #: CR144632 (1610-0737)
Bookmark and Share
  Featured Reviewer  
 
Deduction And Theorem Proving (I.2.3 )
 
 
Knowledge Representation Formalisms And Methods (I.2.4 )
 
 
Learning (I.2.6 )
 
Would you recommend this review?
yes
no
Other reviews under "Deduction And Theorem Proving": Date
Noninteractive zero-knowledge
Blum M., De Santis A., Micali S., Persiano G. SIAM Journal on Computing 20(6): 1084-1118, 1991. Type: Article
Jan 1 1993
Cut elimination and automatic proof procedures
Zhang W. Theoretical Computer Science 91(2): 265-284, 1991. Type: Article
Apr 1 1993
A non-reified temporal logic
Bacchus F., Tenenberg J., Koomen J. Artificial Intelligence 52(1): 87-108, 1991. Type: Article
Oct 1 1992
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy