Abstract: Association Rule is a vital technique applied in the task of Data Mining to search and generate associations among large number of databases. A variety of algorithms have been introduced and used to find the relationships existing in databases. A very popular algorithm is Apriori algorithm. Though it is a very efficient process, it takes a considerable time to scan large amount of transactions in database to generate frequent item sets. Therefore an attempt is made to create an Improved Apriori algorithm that reduces the amount of scans needed to give away the frequent sets.

Keywords: Association Rule, Frequent Item Sets, Original Apriori algorithm, Improved Apriori algorithm.