Utility Pattern Mining Based on the Rule of Utility Leverage

Utility pattern mining is one of the important problem in the field of data mining. However, the current utility mining techniques is to discover high utility patterns, that is, the pattern whose utility exceeds the user-specified threshold, but does not further explore the correlation between different patterns. Therefore, this paper introduces the concept of confidence and lift in high-utility pattern mining, and proposes the conception of Utility Leverage Pattern. In order to effectively discover relevant profitable itemsets, an algorithm called ULPM (Utility Leverage pattern mining) for mining utility pattern based on the utility leverage rule is proposed. The experimental results show that the ULPM algorithm is efficient in mining utility leverage patterns.