A Gleam of Light on Association Rule Mining and Reduction Techniques

Association Rule Mining (ARM) algorithms generate an exceptionally large number of association rules, often in thousands or even millions. Further, the association rules are sometimes very extensive. The nearly impossible task of the end - users to comprehend or validate such large number of complex association rules and limits the usefulness of outcome of mining on massive data sources. Various strategies have been proposed to reduce the number of association rules. They include generating only “non-redundant” rules, generating only “interesting” rules, or generating only those rules satisfying certain other criteria such as coverage, leverage, lift or strength or pruning out ‘irrelevant’ rules. This paper presents the various methods used to solve the issue of association rule reduction.