【作者】 赖桃桃； 冯少荣； 张东站；

【Author】 LAI Tao-tao FENG Shao-rong ZHANG Dong-zhan (Department of Computer Science,Xiamen University,Xiamen 361005,China)

【机构】 厦门大学计算机科学系；

【摘要】 聚类是数据挖掘的重要组成部分;基于划分的聚类算法有近线性时间复杂度的优点,基于密度的聚类算法能发现任意形状的聚类。提出基于划分和密度的聚类算法CUPD(clustering using partition and density),它首先用基于划分的算法对数据进行分类,然后用基于密度的算法进行类的合并。理论分析和实验证明了CUPD算法保持了基于划分和基于密度聚类算法的上述两个优点。更多还原

【Abstract】 Clustering is an important part of the Data Mining.The clustering algorithm based on partition has the advantage that nearly linear time complexity.The density-based clustering algorithm can discover clusters of arbitrary shape.The paper presented a new clustering algorithm CUPD(clustering used partition and density) based on the partition and density.First,using the algorithm based on the partition to classify the data,and then use density-algorithms merger the classes.Theoretical analysis and experimental proofed that the CUPD algorithm in possession of two advantages that algorithms based on the partition and density-clustering have.更多还原