Adaptive categorization in complex systems

A fast and reliable method for categorization of patterns that may be encountered in complex systems is described. Most pattern recognition and classification approaches are founded on discovering the connections and similarities between the members of each class. In this work, a different view of classification is presented. The classification is based on identification of distinctive features of patterns. It will be shown that the members of different classes have different values for some or all of such features. The paper will also show that by making use of the distinctive features and their corresponding values, classification of all patterns, even for complex systems, can be accomplished. The classification process does not rely on any heuristic rules. In this process, patterns are grouped together in such a way that their distinctive features can be explored. Such features are then used for identification purposes.