A distributed framework for trimmed Kernel k-Means clustering

作者：

Highlights：

• Kernel k-Means extension.

• Improved performance over baseline Kernel k-Means.

• Reduced kernel matrix size to make clustering big datasets possible.

• Distributed versions of all assorted algorithms is provided.

• Competitive with the state of the art.

摘要

Highlights•Kernel k-Means extension.•Improved performance over baseline Kernel k-Means.•Reduced kernel matrix size to make clustering big datasets possible.•Distributed versions of all assorted algorithms is provided.•Competitive with the state of the art.

论文关键词：Data clustering,Face clustering,Kernel k-Means,Distributed computing

论文评审过程：Received 9 May 2014, Revised 16 February 2015, Accepted 19 February 2015, Available online 27 February 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.02.020