计算机应用

• 数据挖掘 • 上一篇    下一篇

基于网格熵的边界点检测算法

邱保志 刘洋 陈本华   

  1. 郑州大学信息工程学院 郑州大学信息工程学院
  • 收稿日期:2007-09-26 修回日期:1900-01-01 发布日期:2008-03-01 出版日期:2008-03-01
  • 通讯作者: 刘洋

Grid-entropy-based boundary points detecting algorithm

<a href="http://www.joca.cn/EN/article/advancedSearchResult.do?searchSQL=(((Bao-Zhi QIU[Author]) AND 1[Journal]) AND year[Order])" target="_blank">Bao-Zhi QIU</a> <a href="http://www.joca.cn/EN/article/advancedSearchResult.do?searchSQL=(((Yang LIU[Author]) AND 1[Journal]) AND year[Order])" target="_blank">Yang LIU</a>   

  • Received:2007-09-26 Revised:1900-01-01 Online:2008-03-01 Published:2008-03-01
  • Contact: Yang LIU

摘要: 为了快速有效地检测聚类的边界点,提出了网格熵的概念和基于网格熵的边界点检测算法Greb。该算法利用网格熵的大小来判定聚类的边界点,且只对数据集进行两遍扫描。实验结果表明,对含有任意形状、不同大小以及不同密度且带有噪声的数据集,该算法能快速有效地检测出聚类的边界点。

关键词: 边界点, 网格熵, 聚类

Abstract: In order to detect the boundary points of clusters effectively, the concept of grid-entropy and a grid-entropy-based boundary points detecting algorithm Greb were proposed, which detected boundary points by the value of grid-entropy and only needed to scan the datasets twice. As shown in the experimental results, Greb can detect boundary points effectively and efficiently on various datasets.

Key words: boundary points, grid-entropy, clusters