2011•Advanced materials researchRequires access
A New Algorithm of Association Rules Mining Based on Relation Matrix
Ping Shui Wang
Open publisher page 1 citations
Abstract
Association rule mining is one of the hottest research areas that investigate the automatic extraction of previously unknown patterns or rules from large amounts of data. Finding association rules can be derived based on mining large frequent candidate sets. Aiming at the poor efficiency of the classical Apriori algorithm which frequently scans the business database, studying the existing association rules mining algorithms, we proposed a new algorithm of association rules mining based on relation matrix. Theoretical analysis and experimental results show that the proposed algorithm is efficient and practical.