Association rules mining using evolutionary algorithms
This paper deals with association rules mining using evolutionary algorithms. All previous bio-inspired based association rules mining approaches generate non admissible rules which the end-user can not exploit them. In this paper, we propose two approaches permit to avoid non admissible rules by developing new strategy called delete and decomposition strategy. If an item is appeared in the antecedent and the consequent parts of given rule, this rule is composed on two admissible rules. Then, we delete such item to the antecedent part of the first rule and we delete the same item to the consequent part of the second rule. We also proposed two approaches (IARMGA and IARMMA), the first approach uses a classical genetic algorithm in the search process. However, the second one employs a mimetic algorithm to improve the quality of returned rules. To demonstrate the suggested approaches, several experiments have been carried out using both synthetic and reals instances. The results reveal that it has a compromise between the execution time and the quality of output rules. Indeed, IARMGA is faster than IARMMA whereas the last one outperforms IARMGA in terms of rules quality.
Association rules mining, Genetic Algorithm, Mimetic Algorithm, Inadmissible Rules