Enhancing in-tree-based clustering via distance ensemble and kernelization

作者：

Highlights：

• A distance-ensemble-based framework is proposed for the in-tree-based clustering.

• The in-tree is kernelizd for handling the extremely linearly-inseparable clusters.

• The new in-tree-based clustering method achieves striking robustness and accuracy.

• This paper also derives a new effective and robust in-tree-based distance metric.

摘要

•A distance-ensemble-based framework is proposed for the in-tree-based clustering.•The in-tree is kernelizd for handling the extremely linearly-inseparable clusters.•The new in-tree-based clustering method achieves striking robustness and accuracy.•This paper also derives a new effective and robust in-tree-based distance metric.

论文关键词：In-tree,Distance ensemble,Kernelization,Clustering

论文评审过程：Received 27 February 2019, Revised 19 May 2020, Accepted 29 October 2020, Available online 29 October 2020, Version of Record 30 January 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107731