Abstract
Association rule mining is a data mining technique used to find interesting associations between items in a transaction database. Well known algorithms for association rule mining include Apriori and FP-tree. Apriori is a level wise algorithm and works by scanning the database multiple times. In an attempt to optimize the performance of Apriori, many variations of basic Apriori algorithm have been proposed. These variations exploit different approaches including reducing the number of database scans performed, using special data structures, using bitmaps and granular computing. Majority of these approaches are improvements in implementation of the same basic algorithm. In this paper we propose the RSO (Reduced Set Operations) algorithm, based on reducing the number of set operations performed. RSO is an algorithmic improvement to the basic Apriori algorithm; it is not an implementation improvement. Our analysis shows that RSO is asymptotically faster than Apriori. Experimental results also validate the efficiency of our algorithm.KeywordsAssociation RuleFrequent ItemsetsAssociation Rule MiningApriori AlgorithmGranular ComputingThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call