The Imperialist Competitive Algorithm for Automated Mining of Association Rules-

 Association rule mining is an optimization problem because of several limitations. Recently, the imperialist competitive algorithm (ICA) has been submitted for solving different optimization problems. This algorithm is based on the socio-political competition among empires. This paper proposes a novel ICA algorithm for automated mining of the exciting and readable association rules without considering the minimum support and confidence thresholds. The convergence rate and computational efficiency of ICA have been improved. This study shows that ICA is combined with some operators of genetic algorithms. The experimental results show that this algorithm is more efficient than the methods of mining association rules based on the basic ICA and the genetic algorithm. These modifications are not only valid for association rule mining but also have extensions to other optimization problems.