Site Map - Computational Intelligence

 

The IEEE World Congress on Computational Intelligence that was held in Orlando, Florida in the summer of 1994 provided, for the first time, an academic forum that facilitated the discussion and long debates on the concepts of neural networks, fuzzy systems, evolutionary computation and generic algorithms. Toward the end of the Congress almost everybody began to realise the significance of the term "Computational Intelligence" as the bond that keeps under the same umbrella and facilitates synergistic collaborations. Most recently, the IEEE International Conference on Neural Networks (ICNN) and these disciplines Information Conference on Evolutionary Computing Conferences (ICEC) held jointly in Perth, became another example indicating the necessity for providing such academic forums.

The development of useful intelligent systems has always been a goal for scientists and engineers. Artificial Intelligence has been providing a shed to the above efforts for many years. However, Artificial Intelligence is traditionally concerned with symbolic manipulation and rule-based systems. Although advanced practical systems have been constructed based on these traditional methods, the incentive to continue along these lines can be enhanced by the development of stronger links to human intelligence. It is anticipated that the proper connection to human intelligence will maintain the focus and inspiration for building even "smarter" systems.

Computational Intelligence (CI) has been used in this special area for providing solutions or alternative methodologies. It has been often claimed that one of the major research areas in CI is the modeling of human problem-solving and decision making. Some of the components of this field that are related to decision making are, for example, the acquisition of knowledge, the decision making strategy, the means of inserting new knowledge or modifying previous knowledge, and the development of a friendly user interface. It has been lately demonstrated that neural networks, genetic algorithms, and fuzzy systems can greatly help in this direction. What turns these new propositions into promising tools are their natural fault-tolerance properties, and their capability of leading, from limited or incomplete data, to near-optimum solutions. In this context, novel propositions are introduced as tools for building intelligent systems. Needless to say, such systems do not mean to replace the expert from being the decision maker but, rather, they attempt to enhance the one's abilities to reach a correct decision.