Mining High Utility Pattern in One Phase without Candidate Generation using up Growth+ Algorithm

Utility mining developed to address the limitation of frequent itemset mining by introducing interestingness measures that satisfies both the statistical significance and the user’s expectation. Existing high utility itemsets mining algorithms two steps: first, generate a large number of candidate itemsets and second, identify high utility itemsets from the candidates by an additional scan of the original transaction database. The performance holdup of these algorithms is the generate more no of candidates itemsets and increasing of the number of long transaction itemsets it cannot work minimum utility threshold, the situation may become worse and also creating more no tree. To overcome these problems, propose an efficient algorithm, namely UP-Growth (Utility Pattern Growth), for mining high utility itemsets with pruning techniques for pruning candidate itemsets. The information of high utility itemsets is stored in a special data structure named UP-Tree (Utility Pattern Tree) such that the candidate itemsets can be generated with only two scans of the database. The performance of UP growth+ was evaluated in comparison with the state-of-the-art algorithms on different types of datasets. The experimental results show that UP growth+ outperforms other algorithms in terms of both execution time and memory space under minimum utility threshold is, the more observable its advantage will be it can achieve the level of about two orders of magnitude faster than the state-of-theart algorithms on dense dataset, and more than one order of magnitude on sparse datasets. Keywords—Utility Pattern Growth, UP Tree, High Utility mining, reducing search space, Pruning.