A new index for clustering validation with overlapped clusters
作者：
Highlights：
• An index to compare clustering solutions with overlapped groups is proposed.
• The index is carefully designed with an intuitive probabilistic approach.
• Results with standard datasets for benchmarking are included.
• It has been applied also to a real application involving social networks.
• The index can measure correctly the similarity between clustering solutions.
摘要
•An index to compare clustering solutions with overlapped groups is proposed.•The index is carefully designed with an intuitive probabilistic approach.•Results with standard datasets for benchmarking are included.•It has been applied also to a real application involving social networks.•The index can measure correctly the similarity between clustering solutions.
论文关键词：Overlapped clusters,Validation index,External validation,Cluster perturbation
论文评审过程：Received 24 February 2016, Revised 15 July 2016, Accepted 3 August 2016, Available online 8 August 2016, Version of Record 10 August 2016.
论文官网地址：https://doi.org/10.1016/j.eswa.2016.08.021
