计算机应用 ›› 2005, Vol. 25 ›› Issue (08): 1818-1820.DOI: 10.3724/SP.J.1087.2005.01818

• 数据库与人工智能 • 上一篇    下一篇

一种改进的带障碍的基于密度和网格的聚类算法

严馨1,周丽华2,陈克平2,徐广义3   

  1. 1.昆明理工大学计算机科学技术系; 2.云南大学计算机科学与工程系; 3.南天电子信息股份有限公司
  • 发布日期:2011-04-07 出版日期:2005-08-01
  • 基金资助:

    国家自然科学基金资助项目(60463004);;云南省教育厅科学研究基金资助项目(03Y173D)

Improved clustering algorithm based on density and grid in the presence of obstacles

YAN Xin1,ZHOU Li-hua2,CHEN Ke-ping2,XU Guang-yi3   

  1. 1. Department of Computer Science and Technology,Kunming University of Science and Technology, Kunming Yunnan 650051, China; 2. Department of Computer Science and Engineering, Yunnan University, Kunming Yunnan 650091, China; 3. Nantian Electronics Information Corporation Limited, Kunming Yunnan 650041, China
  • Online:2011-04-07 Published:2005-08-01

摘要: 提出了一个改进的带障碍的网格弥散聚类算法DCellO1:以网格为基础,将基于密度的聚类算法与图形学种子填充算法相结合。该算法能进行任意形状的带障碍聚类,并且在对象分布不均匀时也能获得较好的聚类结果。实验证明了该算法的有效性与优越性。

关键词: 聚类, 网格, 密度, 障碍

Abstract: An improved grid diffusant clustering algorithm in the presence of obstacles called DCellO1 was proposed. Based on grid,it combined density-based clustering algorithm with seed-filling algorithm of graphics. It could construct arbitrary shape clustering in the presence of obstacles, and could obtain good clustering results when the objects distributed unevenly. The experiments prove the superiority and effectiveness of DCellO1.

Key words: clustering, grid, density, obstacles

中图分类号: