| 基于MapReduce的ROCK聚类算法 |
| |
| 引用本文： | 赵雪,;陈龙飞.基于MapReduce的ROCK聚类算法[J].河北职业技术师范学院学报,2014(1):26-32. |
| |
| 作者姓名： | 赵雪 ;陈龙飞 |
| |
| 作者单位： | [1]燕山大学信息科学与工程学院,河北秦皇岛066004; [2]河北经贸大学信息技术学院,河北秦皇岛066004; |
| |
| 基金项目： | 河北省社会科学基金项目（项目编号：HB12GL023）. |
| |
| 摘 要： | 随着大数据时代的到来,聚类分析算法将面临如数据量巨大、数据维数增加等挑战,分布式处理是解决这类问题的方法之一.本研究将ROCK算法与Hadoop平台相结合,按照分布式处理原则,通过计算机集群模式去处理大规模的多样性数据.实验证明,在Hadoop平台下的ROCK聚类算法很大程度上提升了对高维数据进行聚类的能力.
|
| 关 键 词： | MapReduce Hadoop ROCK聚类算法 |
The ROCK Clustering Algorithm Based on MapReduce |
| |
| Institution: | ZHAO Xue , CHEN Long-fei (1 Information Technology and Engineering College, Yanshan University,Qinhuangdao Hebei ,066004; 2 Information Technology College, Hebei University of Economics and Business;China) |
| |
| Abstract: | With the advent of big data era, clustering algorithms have to face new challenges, such as data amounts, data dimensions, etc. distributed processing is one way to solve such problems. ROCK algorithm is combined with Hadoop platform to process the data via computer clusters form in accordance with the distribu- ted processing principles. Experiments show that the algorithms enhance the ability of high-dimensional data of clustering largely. |
| |
| Keywords: | MapReduee Hadoop ROCK Clustering algorithm |
| 本文献已被 维普 等数据库收录！ |
|
