Recent developments in natural computation

The main objective of this special issue is to present readers with the significantly extended and improved versions of several high-quality articles presented at the Third International Conference on Natural Computation (ICNC’07). ICNC’07 was held jointly with The Fourth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD’07) from August 24 to 27, 2007 in Haikou, Hainan, China. ICNC’07 featured the most upto-date research results in computational paradigms inspired from nature, including biological, ecological, linguistic, and physical systems. It is an exciting and emerging interdisciplinary area in which a wide range of techniques and methods are being studied for dealing with large, complex, and dynamic problems. It received 1752 submissions from 35 countries/regions. After rigorous reviews, 770 papers were published in the ICNC’07 proceedings. The acceptance rate for publication was 44%. Through a careful selection from the papers in the ICNC’07 proceedings published by IEEE Press, authors of 18 papers were invited to submit extended versions to this special issue. All authors except one submitted their version to Elsevier Editorial System for Neurocomputing. Each paper had gone though reviewing by at least three reviewers. Based on reviewers’ comments, authors made further revisions. We present 12 papers in this special issue. The first paper entitled ‘‘Behavioral task processing for cognitive robots using artificial emotions’’ authored by Evren Daglarli, Hakan Temeltas, and Murat Yesiloglu presents an artificial emotional-cognitive system-based autonomous robot control architecture. There are three levels, namely, behavioral system level, behavioral selection level, and emotion–motivation module, in its general control architecture. ‘‘Binary classification using ensemble neural networks and interval neutrosophic sets’’ by Pawalai Kraipeerapun and Chun Che Fung proposes two approaches for binary classification. The first one applies a single pair of neural networks while the second one uses an ensemble of pairs of neural networks for the binary classification. Interval neutrosophic sets are used in both methods in order to represent imperfection in the prediction. ‘‘Identification and control of nonlinear systems by a timedelay recurrent neural network’’ by Hongwei Ge, Wenli Du, Feng Qian, and Yanchun Liang introduces time-delay and recurrent mechanisms into a recurrent neural network. A dynamic recurrent back-propagation algorithm is developed for training the proposed neural network. Experiments show that the proposed network is very effective for identification and control for dynamic systems.