Top-K high utility item set identification in big data by MUP-growth the evolutionary approach with less time constraints

High utility itemset mining is a eminent data mining technique used for acquiring the itemsets with high utility among the transactional dataset. As it supports various proposed analysis, it is adopted in a distinct domain applications, ranging from network to medical records data. At present there is huge amount of data generation from various sources, different algorithms have been promoted to handle such a data and also used to recognize high utility itemsets. This research, evaluates MUP-Growth (Multithreaded Utility pattern growth) algorithm to address the high utility itemset mining problem in big data domain with minimum amount of time constraints. The information of such a high utility itemset is maintained in tree data structure known as UP-Tree(utility pattern tree). In this paper, we propose a new framework for mining top-k high utility itemset, where k is the desired number of HUIs to be mined. Performance of proposed algorithm is computed on different datasets and compared with previous approach. Experimental evaluation shows that proposed algorithm out performs better in terms of time constraints. Finally, based on the research, it gives forthcoming research direction to expand any application in the region of pattern mining by selecting the proper combination of these technologies.