A Survey of Distributed Clustering Algorithms

Abstract

Clustering is to divide a set of objects into multiple classes, and each class is made up of similar objects. Traditional centralized clustering algorithms cluster objects stored in a single site, but it cannot satisfy the clustering requirements when objects are distributed. Distributed clustering algorithms can satisfy this need, which extracts a classification mode from distributed objects. This paper classifies and analyzes typical distributed clustering algorithms. Two data sets-Iris and Wine are used to compare several distributed clustering algorithms from two metrics: clustering accuracy and clustering time.