Efficient High Utility Top-K Frequent Pattern Mining from High Dimensional Datasets

Abstract

High utility pattern mining can be defined as discovering sets of patterns that not only co-occurs but they carry high profit. In two-phase pattern mining an apriori algorithm is used for candidate generation. However candidate generation is costly and it is challenging problem that if number of candidate are huge then scalability and efficiency are bottleneck problems. As a rule, finding a fitting least utility edge by experimentation is a monotonous procedure for clients. In the event that min_util is set too low, an excessive number of HUIs will be produced, which may bring about the mining procedure to be exceptionally wasteful. Then again, if min_util is set too high, it is likely that no HUIs will be found. In this paper, we address the above issues by proposing another structure for top-k high utility thing set mining, where k is the coveted number of HUIs to be mined. Two sorts of proficient calculations named TKU (mining Top-K Utility thing sets) and TKO (mining Top-K utility thing sets in one stage) are proposed for mining such thing sets without the need to set min_util. We give an auxiliary examination of the two calculations with talks on their preferences and restrictions. Exact assessments on both genuine and manufactured datasets demonstrate that the execution of the proposed calculations is near that of the ideal instance of best in class utility mining calculations.