• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of Computers >A Parallel Attribute Reduction Algorithm based on Affinity Propagation Clustering
【24h】

A Parallel Attribute Reduction Algorithm based on Affinity Propagation Clustering

机译：基于关联传播聚类的并行属性缩减算法

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

—As information technology is developing rapidly, massive and high dimensional data sets have appeared in abundance. The existing attribute reduction methods are encountering bottleneck problem of timeliness and spatiality. AP(Affinity Propagation) is an efficient and fast clustering algorithm for large dataset compared with the existing clustering algorithms. This paper discusses attribute clustering method in order to reduce attributes and provides a kind of parallel attribute reduction algorithm based on Affinity Propagation （APPAR） clustering. The attribute set is clustered into several subsets by Affinity Propagation algorithm first, and then the reductions of these subsets are proposed concurrently in order to get attribute reduction set of the whole data set. The whole algorithm has been improved in the two sides so as to largely increase the algorithm’s speed. Experimental results show that the APPAR method is outperforming traditional attribute reduction algorithm for huge and high dimensional dataset processing.
机译：-AS信息技术正在迅速发展，大量和高维数据集出现在丰度中。现有的属性还原方法遇到了及时性和空间性的瓶颈问题。 AP（亲和力传播）是与现有聚类算法相比的大型数据集的有效且快速的聚类算法。本文讨论了属性群集方法，以减少属性并提供一种基于关联传播（Appar）群集的并行属性缩减算法。该属性集首先通过关联传播算法群集多个子集，然后同时提出这些子集的缩短，以便获得整个数据集的属性减少集。两侧的整个算法已得到改善，从而大大提高了算法的速度。实验结果表明，该临床方法优于巨大和高维数据集处理的传统属性抑制算法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号