摘要
CURE算法是一种凝聚的层次聚类算法,它首先提出了使用多代表点描述簇的思想。本文通过对已有的基于多代表点的层次聚类算法特点的分析,提出了一种新的基于多代表点的层次聚类算法WRPC。它使用了基于影响因子的簇代表点选取机制和基于k-近邻方法的小簇合并机制,可以发现形状、尺寸更为复杂的簇。实验结果表明,该算法在保证执行效率的情况下取得了更好的聚类效果。
As an agglomerative hierarchical clustering algorithm, CURE firstly employs the method of representing clusters by selecting some 'representative points'. Through the analysis of the feature of traditional hierarchical clus- tering algorithm, a novel agglomerative hierarchical clustering algorithm called WRPC is proposed in this paper. WR- PC can identify clusters with complex shapes and avrious size by introducing the influence-weight-based representative points selection mechanism and k-nearest-neighbor-method-based clusters nesting mechanim. Experimental results show that WRPC can provide better clustering result with high executing efficiency.
出处
《计算机科学》
CSCD
北大核心
2005年第5期150-154,共5页
Computer Science
基金
本文得到教育部重点科学技术研究项目(02038)
天津自然科学基金(023600611)
南开大学亚洲研究中心资助