Fuzzy Utility Mining Under Minimum Weight Constraint of Multiple Item Weights

In this paper, we extend Lan et al.'s fuzzy utility mining approach to handle different weights for items. To address this issue, an efficient algorithm called the weighted fuzzy utility mining is proposed to efficiently discover high weighted fuzzy utility itemsets. However, the weighted fuzzy utility mining problem cannot hold the downward-closure property. We thus design an upper-bound model to prevent information loss in mining. Based on this model, we suggest a two-stage method for weighted fuzzy utility mining. Finally, some experiments are conducted to show the performance of the proposed approach.