A Parallel and Distributed Method to mine Erasable Itemsets from High utility patterns

High utility pattern mining becomes a very important research issue in data mining by considering the non-binary frequency values of items in transactions and different profit values for each item. These profit values can be computed efficiently inorder to determine the gain of an itemset which in-turn will help in production planning of any company. This gain value is needed to prune some of the irrelevant items from the high utility patterns at the time of economic crisis through erasable itemset mining. But all of the existing erasable itemset mining algorithms are based on centralized database and today’s internet era databases are large and inherently distributed. This inherent distribution source of data and the voluminous in size emerges to develop scalable parallel and distributed approach for erasable itemset mining. This paper proposes a parallel method in distributed environment through which the erasable itemsets can be mined from the high utility patterns without which the loss of the profit is no more than the given threshold. This algorithm is designed in such a way so that it can efficiently compute the gain of an itemset as well as prune the irrelevant data or items from the high utility patterns and decrease the execution time.