Damped window based high average utility pattern mining over data streams
作者：
Highlights：
• We suggest a novel algorithm of recent high average utility pattern mining based on the damped window model.
• Novel data structures and pruning techniques are devised to prevent useless mining operations.
• Various experiments are conducted with actual medical datasets composed of various attributes.
• We show that the proposed algorithm has better performance in terms of runtime, memory, scalability and significant test.
• We analyze significant medical patterns mined from actual medical dataset for showing the usefulness of our approach.
摘要
•We suggest a novel algorithm of recent high average utility pattern mining based on the damped window model.•Novel data structures and pruning techniques are devised to prevent useless mining operations.•Various experiments are conducted with actual medical datasets composed of various attributes.•We show that the proposed algorithm has better performance in terms of runtime, memory, scalability and significant test.•We analyze significant medical patterns mined from actual medical dataset for showing the usefulness of our approach.
论文关键词：Data mining,Stream pattern mining,Damped window model,High-average utility,Significant test
论文评审过程：Received 20 April 2017, Revised 18 December 2017, Accepted 27 December 2017, Available online 28 December 2017, Version of Record 14 February 2018.
论文官网地址：https://doi.org/10.1016/j.knosys.2017.12.029
