Incremental Mining of Association Rules by Using Adaptive Algorithm

Abstract

Data mining has recently attracted tremendous amount ofattention in the database research because of its wide applicability in many areas, including decision support, market strategy and financial forecast. One of the most important data mining applications is that of mining association rules. There have been many algorithms on efficient discovery of association rules in large database. However, as the databases grow, the discovered rules need to be verified. It is nontrivial because a database may allow frequent or associational updates and such updates may not only invalidate some existing strong association rules but also turn some existing strong association rules but also turn some weak rules into strong ones. Therefore, mining afresh every time the database grows is needed to be efficient. In this paper, mining association rules for treatment database is implemented by using Adaptive algorithm. Adaptive algorithm make use of large and candidate itemsets and their counts in the older databse, and scan the increment to find which rules continue to prevail and which ones fail in the merged databased. The primary aim of this algorithm is to avoide or minize scans of the older database by using the intermediate data constructucted during the earlier mining.