一种基于相似性度量的高维数据聚类算法的研究

针对传统基于距离度量的聚类算法难以适合高维数据聚类以及高维数据之间相似度难定义的问题,提出了一种新的高维数据聚类算法.该算法基于一个能够更准确地表达出高维对象之间相似性的度量函数,首先计算对象两两之间的相似度并得出一个相似度矩阵,然后根据该相似度矩阵和阈值大小自底向上对数据进行聚类分析.实验结果显示,该算法能够获得质量更高的聚类结果,并且不受孤立点影响,对输入数据顺序也不敏感.

ON CLUSTERING ALGORITHM OF HIGH DIMENSIONAL DATA BASED ON SIMILARITY MEASUREMENT

Aiming at the problems that the traditional clustering algorithms based on distance measurement are not suitable for high dimensional data clustering and the similarities among high dimensional data are hard to be defined,a new high dimensional data clustering algorithm is proposed in this paper.The new algorithm is based on a similarity measuring function,which is able to more accurately express the similarity degree among high dimensional data.The executing progress of the algorithm is as follows:firstly ...