DBCURE-MR: An efficient density-based clustering algorithm for large data using MapReduce
作者：
Highlights：
• A density-based clustering algorithm DBCURE can find clusters with varying densities.
• DBCURE is a generalization of DBSCAN using ellipsoidal neighborhoods.
• We propose a parallel version of DBCURE, called DBCURE-MR, using MapReduce.
• DBCURE-MR finds clusters correctly based on the definition of density-based clusters.
• Experimental results show the efficiency and scalability of the proposed algorithms.
摘要
Highlights•A density-based clustering algorithm DBCURE can find clusters with varying densities.•DBCURE is a generalization of DBSCAN using ellipsoidal neighborhoods.•We propose a parallel version of DBCURE, called DBCURE-MR, using MapReduce.•DBCURE-MR finds clusters correctly based on the definition of density-based clusters.•Experimental results show the efficiency and scalability of the proposed algorithms.
论文关键词：Clustering algorithm,Density-based clustering,Parallel algorithm,MapReduce
论文评审过程：Received 23 April 2013, Revised 18 November 2013, Accepted 22 November 2013, Available online 1 December 2013.
论文官网地址：https://doi.org/10.1016/j.is.2013.11.002
