FSKYMINE: A Faster Algorithm For Mining Skyline Frequent Utility Itemsets
Hung Manh Nguyen, Anh Viet Phan, Lai V Pham · 2019 6th NAFOSTED Conference on Information and Computer Science (NICS) · 2019
Skyline frequent utility itemsets mining is a challenging task in frequent itemsets mining and plays an important role in many data mining applications. Previous studies presented two algorithms, namely SKYMINE and SKYMINE2, to mine Skyline Frequent Utility Itemsets (SFUIs). In which, the SKYMINE2 based on utility-list data structure is a state-of-the-art algorithm. However, the SKYMINE2 remains computationally expensive because the algorithm generates numerous utility lists, join operations, and potential SFUIs. In this paper, we propose a more effective algorithm to mine the SFUIs based on extent utility list data structure and using a new strategy to prune the potential skyline frequent utility itemsets. Notably, experimental results with four datasets show the proposed algorithm reduces the number of utility lists, join operations, potential SFUIs effectively and outperforms the SKYMINE2 in terms of runtime.