Abstract:

As we can see in research world, ant algorithms have showed magnificent research interest within the search/optimization community. In Data Mining search space, Associative Classification [3] (AC) approaches proved to be strongly able to extract more accurate classifiers than traditional classification techniques, In this paper the two are brought together. An investigation of the ant colony algorithms is presented as a variant for associative classification mining. In the proposed algorithm, each class is labeled and assigned to represent the consequent part of the rules within a processor. The ants groups are attached to the main processor to search for the antecedent part of the rules. The ants, depending on the pheromone, and heuristic information, select the values of the attributes according to the importance of an attribute to the class. The proposed algorithm is expected to discover classification rules with magnificent enhancement on accuracy and less redundancy than comparing methods.