SNCStream+: Extending a high quality true anytime data stream clustering algorithm

作者：

Highlights：

• SNCStream+ presents high clustering quality accordingly to the Cluster Mapping Measure.

• SNCStream+ possesses diminished computational complexity when compared to its ancestor.

• SNCStream+ is able to diminish the impact of the curse of dimensionality through the usage of specific distance metrics.

摘要

Highlights•SNCStream+ presents high clustering quality accordingly to the Cluster Mapping Measure.•SNCStream+ possesses diminished computational complexity when compared to its ancestor.•SNCStream+ is able to diminish the impact of the curse of dimensionality through the usage of specific distance metrics.

论文关键词：Data stream clustering,Unsupervised learning,Social networks theory

论文评审过程：Received 27 May 2015, Revised 29 February 2016, Accepted 17 June 2016, Available online 25 June 2016, Version of Record 11 July 2016.

论文官网地址：https://doi.org/10.1016/j.is.2016.06.007