¡@

Utility mining has recently been an important issue due to its wide applications. An itemset in traditional utility mining considers individual profits and quantities of items in transactions regardless of its length. The average-utility measure, which is the total utility of an itemset divided by its number of items within it, was then proposed to reveal a better utility effect than the original utility measure. A mining algorithm was also proposed to find high average-utility itemsets from a transaction database. However, the previous mining approach was based on the principle of level-wise processing to find high average-utility itemsets from a database. In this paper, we thus propose an efficient average-utility mining approach which adopts a projection technique and an indexing mechanism to speed up the execution and reduce the memory requirement in the mining process. The proposed approach can project relevant sub-databases for mining, thus avoiding some unnecessary checking. In addition, a pruning strategy is also designed to reduce the number of unpromising itemsets in mining. Finally, the experimental results on synthetic datasets and two real datasets show the superior performance of the proposed approach.

Keywords: data mining, association rule mining, utility mining, average utility, indexing mechanism

Retrieve PDF document (201201_13.pdf)

Received March 1, 2011; revised August 7, 2011; accepted August 27, 2011.
Communicated by I-Chen Wu.