Association Rule Mining Using Genetic Algorithm: The Role of Estimation Parameters

Genetic Algorithms (GA) have emerged as practical, robust optimization and search methods to generate accurate and reliable Association Rules. The performance of GA for mining association rules greatly depends on the GA parameters namely population size, crossover rate, mutation rate, fitness function adopted and selection method. The objective of this paper is to compare the performance of the Genetic algorithm for association rule mining by varying these parameters. The algorithm when tested on three datasets namely Lenses. Iris and Haberman indicates that the accuracy depends mainly on the fitness function which is the key parameter of GA. The population size is affected by the size of the dataset under study. The crossover probability brings changes in convergence rate with minimal changes in accuracy. The size of the dataset and relationship between its attributes also plays a role in achieving the optimum accuracy.