High Utility Itemset Mining Using Transaction Utility of Itemsets


KIPS Transactions on Software and Data Engineering, Vol. 4, No. 11, pp. 499-508, Nov. 2015
10.3745/KTSDE.2015.4.11.499,   PDF Download:

Abstract

High utility itemset(HUI) mining refers to the discovery of itemsets with high utilities which are not less than a user-specified minimum utility threshold, by considering both the quantities and weight factors of items in a transaction database. Recently the utility-list based HUI mining algorithms have been proposed to avoid numerous candidate itemsets and the algorithms need the costly join operations. In this paper, we propose a new HUI mining algorithm, using the utility-list with additional attributes of transaction utility and common utility of itemsets. The new algorithm decreases the number of join operations and efficiently prunes the search space.Experimental results on both synthetic and real datasets show that the proposed algorithm outperforms other recent algorithms in runtime, especially when datasets are dense or contain many long transactions.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
S. Lee and J. S. Park, "High Utility Itemset Mining Using Transaction Utility of Itemsets," KIPS Transactions on Software and Data Engineering, vol. 4, no. 11, pp. 499-508, 2015. DOI: 10.3745/KTSDE.2015.4.11.499.

[ACM Style]
Serin Lee and Jong Soo Park. 2015. High Utility Itemset Mining Using Transaction Utility of Itemsets. KIPS Transactions on Software and Data Engineering, 4, 11, (2015), 499-508. DOI: 10.3745/KTSDE.2015.4.11.499.