Efficiently Using Prefix-trees in Mining Frequent Itemsets

Efficient algorithms for mining frequent itemsets are crucial for mining association rules. Methods for mining frequent itemsets and for iceberg data cube computation have been implemented using a prefix-tree structure, known as an FP-tree, for storing compressed information about frequent itemsets. Numerous experimental results have demonstrated that these algorithms perform extremely well. In this paper we present a novel array-based technique that greatly reduces the need to traverse FP-trees, thus obtaining significantly improved performance for FPtree based algorithms. Our technique works especially well for sparse datasets. Furthermore, we present new algorithms for a number of common data mining problems. Our algorithms use the FP-tree data structure in combination with our array technique efficiently, and incorporates various optimization techniques. We also present experimental results which show that our methods outperform not only the existing methods that use the FP-tree structure, but also all existing available algorithms in all the common data mining problems.