Mining high average-utility itemsets

The average utility measure is adopted in this paper to reveal a better utility effect of combining several items than the original utility measure. A mining algorithm is then proposed to efficiently find the high average-utility itemsets. It uses the summation of the maximal utility among the items in each transaction including the target itemset as the upper bounds to overestimate the actual average utilities of the itemset and processes it in two phases. As expected, the mined high average-utility itemsets in the proposed way will be fewer than the high utility itemset under the same threshold. Experiments results also show the performance of the proposed algorithm.