Path-scan: A novel clustering algorithm based on core points and connexity

作者：

Highlights：

• The clustering algorithm is oriented for a discovery task

• Clusters are characterized by several core and support patterns

• They can be path-connected via a density differential approach

• The Dijkstra algorithm is applied to produce the optimal path

• Performance is assessed using synthetic and real-world data

摘要

•The clustering algorithm is oriented for a discovery task•Clusters are characterized by several core and support patterns•They can be path-connected via a density differential approach•The Dijkstra algorithm is applied to produce the optimal path•Performance is assessed using synthetic and real-world data

论文关键词：Clustering,Natural cluster,Distance,Density,Neighbors

论文评审过程：Received 24 June 2021, Revised 12 July 2022, Accepted 27 July 2022, Available online 6 August 2022, Version of Record 12 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118316