Abstract

Abstract: Highutility item sets (HUIs) mining is a rising subject in information mining,which aims to finding all item sets having an utility meeting a clientdetermined least utility edge min util. But, setting min util suitably is atroublesome issue for clients. As a rule, finding a fitting least utility edge byexperimentation is a monotonous procedure for clients. In the event that minutil is set too low, an excessive number of HUI swill be produced, which maybring about the mining procedure to be exceptionally wasteful. Then again, ifmin util is set too high, it is likely that no HUIs will be found. We addressthe above issues by redefining the problem of high utility item sets (HUIs)mining by top-k high utility item sets (top-k HUI) mining, where k is the desirednumber of HUIs to be mined. Two algorithms named TKU (mining Top-K Utility itemsets in two stages) and TKO (mining Top-K utility item sets in one stage) areproposed for mining such item sets without the need to set min util. To improvethe performance, we apply pre-evaluation strategy to algorithms.

Keywords:HUI, Utility Mining, High Utility Itemset Mining, Top-k Pattern Mining, Top-kHigh Utility Itemset Mining