Evolutionary multi-objective automatic clustering enhanced with quality metrics and ensemble strategy

作者：

Highlights：

• A more comprehensive investigation on automatic clustering approaches via EMO is presented.

• Two new methods are proposed w.r.t. different encoding schemes.

• The quality metrics are incorporated to guide optimization.

• Determining the final partitioning from Pareto set is improved with clustering ensemble strategies.

摘要

•A more comprehensive investigation on automatic clustering approaches via EMO is presented.•Two new methods are proposed w.r.t. different encoding schemes.•The quality metrics are incorporated to guide optimization.•Determining the final partitioning from Pareto set is improved with clustering ensemble strategies.

论文关键词：Clustering,Evolutionary multi-objective optimization,Cluster validity index,Ensemble method

论文评审过程：Received 20 March 2019, Revised 30 August 2019, Accepted 30 August 2019, Available online 5 September 2019, Version of Record 20 January 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2019.105018