An empirical evaluation of high utility itemset mining algorithms
作者：
Highlights：
• Running time and memory consumption comparison of 10 HUI mining algorithms.
• Comparison tests using 9 real world and 72 synthetic datasets.
• d2HUP and EFIM are the top-2 performers regarding running time.
• d2HUP is fastest when the data is sparse or the average transaction length is large.
• EFIM is the most efficient algorithm regarding memory consumption.
摘要
•Running time and memory consumption comparison of 10 HUI mining algorithms.•Comparison tests using 9 real world and 72 synthetic datasets.•d2HUP and EFIM are the top-2 performers regarding running time.•d2HUP is fastest when the data is sparse or the average transaction length is large.•EFIM is the most efficient algorithm regarding memory consumption.
论文关键词：Itemset mining,High utility itemsets,State-of-the-art high utility itemset mining
论文评审过程：Received 21 August 2017, Revised 10 January 2018, Accepted 3 February 2018, Available online 8 February 2018, Version of Record 16 February 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.02.008
