Improved Association Mining Algorithm for Large Dataset
Tannu Arora, Rahul Yadav · 2011
Mining frequent patterns in transaction databases, time-series databases, and many other kinds of databases has been studied popularly in data mining research. Most of the previous studies adopt an Apriori-like candidate set generation-and-test approach. However, candidate set generation is still costly, especially when there exist a large number of patterns and/or long patterns. Further, dynamic itemset counting algorithm, an extension to Apriori algorithm used to reduce number of scans on the dataset. It was alternative to Apriori Itemset Generation .In this, itemsets are dynamically added and deleted as transactions are read .It relies on the fact that for an itemset to be frequent, all of its subsets must also be frequent, so we only examine those itemsets whose subsets are all frequent. Both Apriori and DIC are based on candidate generation. But frequent-pattern tree (FP-tree) structure, which is an extended prefix-tree structure for storing compressed, crucial information about frequent patterns, mine the frequent itemsets without any candidate generation. Our performance study shows that the FP-growth method is efficient and scalable for mining both long and short frequent patterns, and is about an order of magnitude faster than the Apriori algorithm and also faster than some recently reported new frequent-pattern mining methods. In this paper, we combine the features of both FP-tree and dynamic itemset so that we can mine the frequent itemsets dynamically without any candidate generation. We propose a new algorithm, Dynamic-FP which takes the advantages of both DIC as well as FP-tree. This algorithm will be compared with previous algorithms to give the better performance for large dataset.