Mining event logs for knowledge discovery based on adaptive efficient fuzzy Kohonen clustering network
作者：
Highlights：
• A new clustering method AEFKCN is developed to mine design behavioral patterns in BIM logs.
• A new CVI is proposed to reduce computation complexity and dependencies on cluster centroids.
• AEFKCN is validated in real logs to speed up convergence and generate better clustering results.
• Design productivity can be evaluated objectively and divided into high, medium, and low levels.
• It contributes to data-driven decision making to allocate tasks and accelerate design processes.
摘要
•A new clustering method AEFKCN is developed to mine design behavioral patterns in BIM logs.•A new CVI is proposed to reduce computation complexity and dependencies on cluster centroids.•AEFKCN is validated in real logs to speed up convergence and generate better clustering results.•Design productivity can be evaluated objectively and divided into high, medium, and low levels.•It contributes to data-driven decision making to allocate tasks and accelerate design processes.
论文关键词：Clustering algorithm,Clustering validity index,Event log mining,Behavioral pattern discovery,Decision making
论文评审过程：Received 9 January 2020, Revised 21 June 2020, Accepted 20 September 2020, Available online 23 September 2020, Version of Record 5 October 2020.
论文官网地址：https://doi.org/10.1016/j.knosys.2020.106482
