Support Vector Machines (SVMs) and related kernel methods are currently very active research areas within neural computation and machine learning. Motivated by statistical learning theory they have been successfully applied to numerous tasks within data mining, computer vision, and bioinformatics, for example. SVMs are examples of a broader category of learning approaches which utilize the concept of kernel substitution, thereby making the task of learning more tractable by exploiting an implicit mapping into a high dimensional space. SVMs have many appealing properties such as solving convex quadratic programming problems and they have been found to work very well in practice. The aim of this session is to present new perspectives and new directions in SVM and kernel methods. We seek contributions from different aspects of this topic: theory, implementations, new methodologies, and applications.

Genetic Algorithms GAs, Artificial Neural Networks (ANN) as well as Fuzzy Systems (FS) are omnipresent in almost every intelligent system design. Just to name few, engineering, control, economics and forecasting are some of the scientific fields that enjoy the use of ANN and FS. Unfortunately, the majority of the GAs, ANNs and FS applications are complex and so require a large computational effort to yield useful and practical results. Therefore, dedicated hardware for intelligent techniques computation is becoming a key issue for designers. With the spread of reconfigurable hardware such as DSPs, FPGAs and FPAAs, digital as well as analog hardware implementations of such computation become cost-effective. The focus of this special session will be on all aspects of intelligent embedded hardware. Of special interest are contributions that describe new and efficient hardware architectures and high speed implementations of intelligent systems.

FS and their Application

Fuzzy Systems FS arethe basis of granular computing, and have been developing successfully. Fuzzy Logic has been applied to many research areas, such as control systems, clustering techniques, machine learning, database, etc. This session will focus on latest developments of FS and their Application. Papers are invited for submission on new conributions in the following, (but not limited to) areas: FS techniques, approaches, theoretical developments,  Dynamic FS, Fuzzy based Machine Learning,  Dynamic Fuzzy Reasoning

This Special Session will focus on Soft Computing (SC) techniques for the development of hybrid intelligent systems for pattern recognition, modeling, simulation and control of non-linear dynamical systems. SC techniques at the moment include (at least) Neural Networks, Fuzzy Logic, Genetic Algorithms and probability Theory. Each of these methodologies has strengths and drawbacks and many problems have been manipulated, relying on one of these methodologies. However, many real-world complex industrial problems require the integration of several of these methodologies to really achieve the efficiency and accuracy needed in practice. The Special Session will include applications on the following areas: Robotic Dynamic Systems, Non-Linear Dynamic Plants, and Manufacturing Systems. Topics of interest (not limited to): Successful new applications to real-world problems, of existing soft computing techniques that are found to achieve better results than conventional techniques. In this case, special attention should be given to the metrics used to compare SC techniques with conventional ones, Developments of innovative hybrid methods combining SC techniques and conventional techniques to solve problems related to modeling, simulation, and control of non-linear dynamical systems. In this case, the problems to be considered in these papers may not be as complex as the ones in the previous point, but the authors have to explain very carefully how their proposed method could be used, in the future, to solve real-world problems, Papers considering original research on new SC techniques are also welcome, but the authors would have to make a detailed description of how their proposed approach is compared with other related techniques.

Artificial Life approach

Artificial Life studies are the study of the simulation and synthesis of living systems. Moreover, this science of generalized living and life-like systems provides techniques with a long time of design expertise to learn from and exploit through the example of the evolution of organic life on earth. Increased understanding of the massively successful design diversity, complexity, and adaptability of life is rapidly making inroads into all areas of engineering and the Sciences of the Artificial. Numerous applications of ideas from nature and their generalizations to synthetic conception engineering and science. Target topics in this special session will include, but not  limited to: Applications of Artificial Life, Self- maintaining and evolvable artificial systems, Bio-automata and Bio-clock,  Bio and artificial adaptation aspects, artificial Immune Systems etc.

The Particle Swarm Optimization  has grown over the last decade. Since its introduction in 1995 the Particle Swarm Optimization paradigm has undergone various  improvements. The aim of this special session is to draw together researchers involved in this area to consolidate and share ideas, advancements and problems in this field. High levels of activity have been experienced in the following broad categories: theoretical and primitive analysis; new PSO algorithms; as well as PSO applications. It is hoped that the special session will attract high quality contributions from, but not limited to, the following topics: Initialization Schemes, Comparative Studies, Convergence analysis, Parameter Selection, Hybrid Models, Discrete PSO.

This special session emphasize on recent on-going research in the field of genome informatics, which is an approach to explore a mechanism of life system coded by genes and evolution of creatures and construct intelligent databases. The genome informatics are extended to, medical science, pharmacy, and agriculture; moreover evolutionary computation itself. The development of powerful tools to acquire knowledge and rules from enormous volumes of gene data is a great demand. Evolutionary computation is one of the promising approaches to meet that requirement. It is hoped that the special session will attract high quality contributions from, but not limited to, the following topics: structure of protein, sequential evolution, probabilistic evolution, clustering techniques of huge data banks, prediction of gene structure and function, DNA and hormone and protein relationship.

Genomics focuses on the study of large sets of genes with the goal of understanding collective function, rather than that of individual genes. Such a study is important since cellular activity and its failure in disease result from multivariate activity among cohorts of genes. Very recent research indicates that engineering approaches for prediction, signal processing and AI  are quite well suited for studying this kind of multivariate interaction. The aim of this workshop will be to provide the attendees with a state of the art account of the research that has been accomplished in this field thus far and to make them aware of some of the open research challenges.

With the increasing use of complex and large-scale information systems new approaches to the design and implementation of these systems merit significant investigation and analysis. Multi-agent systems have been mooted as an important means with which to address the development of large and complex information systems, and provide one such approach to the problem above. Agent technology and multi-agent systems have arisen in an exciting and rapidly changing field during the last ten years, emerging from distributed artificial intelligence. In particular, the exponential growth of the internet as an enabling technology for distributed systems has provided an increasingly urgent need for research into issues surrounding the research paradigms considered in this workshop. Indeed, at this exciting interface between a number of fields of research, there are many open questions to be answered, and many problems to be solved. This session will focus on a broad range of issues relating to the design and implementation of agent-based systems. Topics of interest include but are not limited to: agent networks, agent coordination and integration of activities, agents in Internet/E-Commerce systems and applications,  agents in database systems,  agent adaptation and learning, mobility and Security Issues, industrial agent systems and applications, human-agent interaction.

Hidden Markov Models (HMMs) have been most successful in many applications, particularly in signal processing. More recently, HMMs have become a key tool for many applications in genomics and proteomics. The structure and associated algorithms of HMMs will be fully described, and an industrial application to the automated detection of genes will be presented in order to highlight the fact that the algorithms per se are only part of the solution; knowledge- based heuristics learned in the field, are of utmost importance. Concluding, several other applications will be mentioned.

Rough set theory, originated by Z. Pawlak in 1982, is a formal mathematical theory modeling knowledge in terms of equivalence relations. The main advantage of rough set theory is that it does not need any preliminary or additional information about data (like probability in probability theory, grade of membership in fuzzy set theory, etc.). Rough set theory was applied in a number of areas, mostly in data mining. In our session we anticipate some theoretical papers and some papers describing the newest applications of rough set theory to data mining.

Web-based intelligent learning is becoming more effective. Due to the rapid growth of the use of computers in education, as well as the introduction of the World Wide Web (WWW), a large number of Web-based educational applications have been developed and implemented. However, very few of them are pedagogically intelligent and interactive for learning purposes. The principle of AI made computers more useful, as well as intelligent, in order to utilize them in all the fields of human life. The application of AI principles is the next advanced step to a Web-based ITS. Hence, the influence of AI on software technology has considerably increased. As a result, the use of AI techniques in teaching/learning, such as expert systems, simulations and virtual reality, etc, has become a major part in the development of Web-based intelligent authoring systems. AI is an advanced scientific technology that is used for efficient computer-based problem-solving techniques in various disciplines. The important contribution of AI in computer-based education is to provide knowledge-based access to resources. The back history of computerized educational measurement system shows that each generation of educational measurement has shown an increased use of AI and expert systems approaches in order to improve educational measurement activities.

Recently, various sorts of AI theories and techniques have been developed. Some of them have been integrated or shown their relationships. Others have changed their forms. AI techniques are useful and some of them seem to be utilized in the real world systems. However, a lot of AI theories and techniques need applications to be applied. Also, a lot of applications need techniques that can be achieved by AI theories and techniques. Last decade can be thought of as the first decade when IT (Information Technology) was widely recognized. We think AI can contribute to the IT. In this session, we would like to focus on AI techniques toward information systems. Especially, we would like to focus on the following topics. Intelligent information systems,  prediction, belief revision, constraint reasoning, multi-agent, AI techniques for internet, intelligent support systems for handicapped persons