【作者】 刘彤； 孙永香； 张振洪；

【Author】 Liu Tong~1 Sun Yongxiang~2 Zhang Zhenhong~3 (1 Dept of Information,Shandong University of Science and Technology,Tai’an 271019,China; 2 Institute of Information Science and Engineering,Shandong Agricultural University,Tai’an 271018,China; 3 Section of Personnel,Shandong University of Science and Technology,Tai’an 271019 ,China)

【机构】 山东科技大学信息系； 山东农业大学信息科学与工程学院；

【摘要】 在综合分析基于层次的凝聚聚类算法和基于密度的聚类方法的基础上,提出了一种聚类质量更好的混合聚类算法 (HCAD)。它结合簇的密度参数重新定义了各簇之间的距离,采用层次凝聚的过程得到最终的聚类结果。实验表明其可有效地识别不同形状、大小的簇,过滤"噪声"数据,该算法在保证执行效率的情况下取得了更好的聚类结果。更多还原

【Abstract】 A new hybrid clustering algorithm(HCAD)of good quality was put forward on the foundation of synthetically analy- zing agglomerative clustering algorithm based on hierarchy and clustering algorithm based on density.The main ides of our algo- rithm is using the density of clusters to co.mpute the distance between clusters,and the final clustering can be obtained through the process of hierarchical agglomeration.The experimental results show that our algorithm can effectively identify clusters with different shape and size,and less sensitive to the noise,and it can provide better clustering result with high executing efficiency.更多还原