International Conference on Computer and Knowledge Engineering

Home / 14th International Conference on Computer and Knowledge Engineering

An Improved and Accurate Measure for Mining Correlated High-utility Itemsets

Authors :

Amir Masoud Heidari Orojloo¹ Morteza Keshtkaran²

1- School of electrical and computer engineering University of Shiraz Shiraz, Iran 2- School of electrical and computer engineering University of Shiraz Shiraz, Iran

Keywords :

high-utility itemset mining،correlated high-utility itemset mining،correlation measure،Kulczynski measure،imbalanced ratio

Abstract :

This paper introduces an improved version of the Kulczynski measure, a widely used measure for mining correlated high-utility itemsets. This improved measure aims to achieve high accuracy while efficiently extracting itemsets with high utility that are also correlated. Using the Kulczynski measure often results in the generation of imbalanced itemsets in the mining process. In this paper, by considering the imbalance ratio measure in conjunction with the Kulczynski measure, the number of imbalanced itemsets is reduced significantly. To evaluate the performance of the improved measure, the effect of using this measure in comparison with the Kulczynski measure on the standard datasets is considered in the experiments. The results demonstrate the superior accuracy and advantage of using this correlation measure compared to its competitor.