Improving Efficiency of Apriori Algorithm using Cache Database

One of the most popular data mining approach to find frequent itemset in a given transactional dataset is Association rule mining. The important task of Association rule mining is to mine association rules using minimum support value which is specified by the user or can be generated by system itself. In order to calculate minimum support value, every time the complete database has to be scanned for each item in the transaction. This decreases the time complexity of the algorithm. Here we proposed a new algorithm which scan the database once and create a cache database for each transaction using hash map. This cache copy is then used to search for frequent item sets. Due to which the overhead of scaning complete database for each item is reduced, and efficiency is increased. Key word: Apriori, cache database, hash map, scanning time, time complexity.