作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (20): 4-6. doi: 10.3969/j.issn.1000-3428.2008.20.002

• 博士论文 • 上一篇    下一篇

基于K-means算法的高性能空间索引方法

赵 伟1,2,张 姝2,李文辉1   

  1. (1. 吉林大学计算机科学与技术学院,长春 130012;2. 长春工业大学计算机科学与工程学院,长春 130012)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-10-20 发布日期:2008-10-20

High-performance Space Index Method Based on K-means Algorithm

ZHAO Wei1,2, ZHANG Shu2, LI Wen-hui1   

  1. (1. College of Computer Science & Technology, Jilin University, Changchun 130012; 2. School of Computer Science & Engineering, Changchun University of Technology, Changchun 130012)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-10-20 Published:2008-10-20

摘要: 提出基于R-link树的快速空间索引结构,在该结构中引入K-means算法。在K-means中采用均值-标准差确定初始聚类中心,提高了收敛速度。通过距离准则函数来优化K值,避免K值的盲目选取。与R-link相比空间开销代价有时略大,但换取了更高的性能,且数据量越多,索引结构的整体性能越好。

关键词: 空间数据库, R-link树, 四叉树, 空间聚类, 空间索引

Abstract: This paper presents a quick speed spatial index structure based on R-link tree. And K-means algorithm in the structure is used. In K-means algorithm, value-standard deviation is adopted to ascertain the initial clustering centres to improve convergence speed and ultimate K value is ascertained by distance criterion function to make K value most suitable. The structure sometimes consumes more storage than R-link but gains better performance. Furthermore, the more data quantities, the better this kind of structure’s overall performance is.

Key words: spatial database, R-link tree, quad-tree, spatial clustering, spatial index

中图分类号: