Data mining is an important tool in extracting interesting patterns from large datasets to represent knowledge. Association rule mining is one of the important concepts in Data Mining. Frequent itemset Generation is one the step in Association Rule Mining. In order to discover the relationships among the data items in large size of database, the most of the research activities focus on it. This research work is mainly implemented by focusing on the analysis of frequent itemset generation in customer dataset to find out customers buying behavior. The traditional algorithms Apriori and existing algorithm Cluster Based Bit Vector Association Rule Mining (CBVAR) and a proposed algorithms namely Improved Cluster Based Bit Vector Association Rule Mining (ICBVAR) are taken to find the efficiency of the algorithms in terms of its execution time and occupied space. A comparative analysis of the all the algorithms is carried out and the best algorithm is based on its performance that is suggested. From the experimental results, the proposed algorithm ICBVAR is faster and gives high recognition results.

Association Rule Mining, Apriori Algorithm, Cluster Based Bit Vector Association Rule Mining, ICBVAR, Frequent Itemset Generation.