Mining Top-K High Utility Item Sets without Minimum Utility Threshold

Abstract

High utility item sets mining is an emerging topic in data mining. Here refers discovering all item sets having a utility meeting a user specified minimum utility threshold. According to users it is a difficult to set min_util. If min_util is set too low, too many HUIs will be generated, it may cause the mining process very inefficient. If min_util is set too high, may be there is no HUIs. Here the above issues are addressed by proposing a new method for mining top-k high utility itemset. Here k is mentioned as the number of HUIs to be mined. Two algorithms named TKU (mining Top-K utility itemsets) and TKO(mining Top-K Utility itemsets) are proposed for mining. These algorithms mine the itemsets without setting min_util threshold.