In this paper, we present a new method for mining the smallest set of association rules by pruning rules that are redundant. Based on theorems which are presented in section 4, we develop the algorithm for pruning rules directly in generating rules process. We use frequent closed itemsets and their minimal generators to generate rules. The smallest rules set is generated from minimal generators of frequent closed itemset X to X and minimal generators of X to frequent closed itemset Y (where X is the subset of Y). Besides, a hash table is used to check whether the generated rules are redundant or not. Experimental results show that the number of rules which are generated by this method is smaller than that of non-redundant association rules of M. Zaki and that of minimal non-redundant rules of Y. Bastide et al. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Vo, B., & Le, B. (2010). Mining the most generalization association rules. In Studies in Computational Intelligence (Vol. 283, pp. 207–216). https://doi.org/10.1007/978-3-642-12090-9_18
Mendeley helps you to discover research relevant for your work.