Data mining has recently attracted considerable attention from database practitioners and researchers because of its applicability in many areas, such as decision support, market strategy and financial forecasts combing techniques from the fields of machine learning, statistics and database, data mining enables us to find out useful and invaluable information from huge databases. Mining of association rules has received much attention among the various data mining problems. Most algorithms for association rule mining are the variants of the basic Apriori algorithm. One characteristic of these Apriori-based algorithms is that candidate itemsets are generated in rounds, with the size of the itemsets incremented by one per round. The number of database scan required by Apriori-based algorithms thus depends on the size of the largest large itemsets. In this research we proposed a new candidate set generation algorithm, which generates candidate itemsets of multiple sizes at each iteration by taking input as suggested large itemsets.