Rule Based Interaction Technique with Ant Colony Optimization Algorithm

Ant Colony Optimization (ACO) algorithms have been successfully applied to discover a list of classification rules. In general, these algorithms follow a sequential covering strategy, where a single rule is discovered at each iteration of the algorithm in order to build a list of rules. The problem in this case is not coping with the problem of rule interaction, i.e., the result of a rule affects the rules that can be discovered subsequently since the search space is modified due to the removal of examples covered by previous rules. Here, a new sequential covering strategy cAnt-MinerPB for ACO classification algorithm is proposed to mitigate the problem of rule interaction, where the order of the rules is implicitly encoded as pheromone values and the search is guided by the quality of a candidate list of rules. cAnt-MinerPB algorithm, which is the extended version of the Ant-Miner algorithm that handles continuous attributes on-the-fly during the rule construction process. The experiments are conducted using 18 publicly available data sets and results shows that the predictive accuracy obtained by a new ACO classification algorithm implementing the cAnt- MinerPB, statistically significantly higher than the predictive accuracy of state-of-the-art rule induction classification algorithms.