基于MapReduce的KMeans聚类算法的并行化实现

Parallel Implementation of KMeans Clustering Algorithm Based on MapReduce

聚类分析是数据挖掘的重要研究内容之一,在识别数据的内在结构方面具有重要的作用。目前,在大数据环境下,怎样更加有效地使用该方法来提取海量信息中的有价值的数据,是一个重要的研究方向。文章以常见的聚类算法Kmeans为例,介绍该算法的并行化实现。

The clustering analysis was one of the important research contents of data mining and inherent in the identification data structure was an extremely important role. At present,under the environment of big data,how to more effectively use this method to extract the valuable data in the vast amounts of information was an important research direction. Taking the common Kmeans clustering algorithm as an example,authers introduced the algorithm of parallel implementation.