An Efficient TDTR Algorithm for Mining Frequent Itemsets

Research on mining frequent itemsets is one the emerging task in data mining.The purchasing of one product when another product is purchased represents an association rule. Association rules are useful for analyzing the customer behavior. It takes an important part in shopping basket data analysis, clustering. The FP-Growth algorithm is the basic algorithm for mining association rules. This paper presents an efficient algorithm for mining frequent itemsets using Two Dimensional Transactions Reduction(TDTR) approach which reduces the original database(D) transactions to the reduced data base transactions D1 based on the min_sup count. Then for each item it finds the number of transactions that the item present and hence find the largest frequent itemset using the two dimensional approach. Using the largest item set property ,it finds the subset of frequent item sets. Thus TDTR approach reduces the number of scans in the database and hence improve the efficiency & accuracy by finding the number of association rules and reduces time to find the rules.