Single-pass and linear-time k-means clustering based on MapReduce

作者：

Highlights：

• mrk-means is a novel clustering algorithm which is based on MapReduce.

• mrk-means is single-pass and linear-time.

• mrk-means results in clusters that are O(log2k)-competitive to optimal solution.

• mrk-means is both faster and more accurate than Apache Mahout and GraphLab k-means.

摘要

Highlights•mrk-means is a novel clustering algorithm which is based on MapReduce.•mrk-means is single-pass and linear-time.•mrk-means results in clusters that are O(log2k)-competitive to optimal solution.•mrk-means is both faster and more accurate than Apache Mahout and GraphLab k-means.

论文关键词：Distributed k-means,Data clustering,MapReduce-based clustering

论文评审过程：Received 25 May 2014, Accepted 23 February 2016, Available online 4 March 2016, Version of Record 23 March 2016.

论文官网地址：https://doi.org/10.1016/j.is.2016.02.007