PERFORMANCE COMPARISON- OF MULTILEVEL ASSOCIATION -RULE MINING ALGORITHMS

Abstract

With the widespread computerization in business, government, and science, the efficient and effective discovery of interesting information from large databases becomes essential. Data mining or Knowledge Discovery in Database (KDD) emerges as a solution to the data analysis problems faced by many organizations. Association rule mining finds interesting association among a large set of data items. Mining of association rules mainly focuses at a single conceptual level. In a large database of transaction, where each transaction consist of a set of items, and a taxonomy (is-a hierarchy) on items, it is required to find out the associations at multiple conceptual levels. Mining association rules at multilevel may lead to the discovery of more specific and concrete knowledge from data. In this dissertation, multilevel level association rule mining algorithms have been evaluated and compared. An algorithm has been extended for the cross level association rule mining which discover the additional strong association rules in taxonomy. All * algorithms have been implemented and tested on Synthetic databases which are generated using a randomized algorithm. The performance indices used for performance comparisons are minimum support threshold at different levels and varying number of transactions. All algorithms are implemented" using JAVA language and are tested on Microsoft Windows XP platform.