Estimating the number of clusters in a dataset via consensus clustering

作者：

Highlights：

• We provide a consensus-based approach for the number of clusters estimation problem.

• The method is tested for 21 benchmark datasets.

• The method performs better than other clustering-based approaches (k-means).

• The method performs better than consensus-based approaches.

摘要

•We provide a consensus-based approach for the number of clusters estimation problem.•The method is tested for 21 benchmark datasets.•The method performs better than other clustering-based approaches (k-means).•The method performs better than consensus-based approaches.

论文关键词：Weighted consensus clustering,Validity indices,Number of clusters

论文评审过程：Received 20 November 2018, Revised 9 January 2019, Accepted 29 January 2019, Available online 30 January 2019, Version of Record 4 February 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.01.074