New internal index for clustering validation based on graphs
作者：
Highlights：
• Minimum spanning tree graph-based clustering is applied.
• The concept ``cohesion" is introduced to measure the clusters quality and distance.
• A new internal validation index that captures the geometry of cluster is defined.
• It has been proved effective on 16 datasets of different shapes and sizes.
• The two versions of the new index show better performance than traditional ones.
摘要
•Minimum spanning tree graph-based clustering is applied.•The concept ``cohesion" is introduced to measure the clusters quality and distance.•A new internal validation index that captures the geometry of cluster is defined.•It has been proved effective on 16 datasets of different shapes and sizes.•The two versions of the new index show better performance than traditional ones.
论文关键词：Graphs-based clustering,Internal clustering validation indices,Cluster geometry,Arbitrary shaped clusters,Minimum spanning tree
论文评审过程：Received 10 February 2017, Revised 1 June 2017, Accepted 2 June 2017, Available online 3 June 2017, Version of Record 9 June 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.06.003
