一种基于网格密度的自适应聚类分析算法  

Self-adapted clustering algorithm based on grid density

作　　者：董琰[1] 葛君伟[1] 

机构地区：[1]重庆邮电大学中韩GIS研究所,重庆400065

出　　处：《计算机应用研究》2007年第8期56-57,66,共3页Application Research of Computers

基　　金：国家"863"计划资助项目(2005AA123730)

摘　　要：在结合基于密度和基于网格的聚类算法优点的基础上,提出一种新的聚类算法。该算法能够在海量、高纬数据下发现任意形状的聚类并对噪声数据不敏感,具有较低的时间和空间复杂性及较高的识别率。通过实验对该算法进行了性能比较和测试,显示了它在各方面的优越性。This paper presented a new efficient clustering algorithm that combined the approach based on density and grid. The most creativity of this novel algorithm was capturing the shape and extent of a cluster by using grid, and then analyzed the data based on the grid density. It also could reach high efficiency because of its linear time complexity. Both theory analysis and experimental results prove that this algorithm can discover clusters with arbitrary shape and is insensitive to noise data.

关 键 词：聚类 密度 网格 连通性 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]