【摘要】 高维数据聚类问题是当前聚类分析研究的重点.笔者提出了一种改进的基于超网络的高维数据聚类算法.首先,将高维数据映射到一个大规模带权超网络中;其次,定义超网络中边的权重;再次,采用优化的超图划分方法 划分带权超网络;最后实现高维数据聚类.这样有效过滤掉聚类中的噪声数据,避免了传统聚类方法 在降维过程中产生的弊端.实验证明,该算法具有较理想的有效性和精确度.更多还原

【Abstract】 High- dimensional data clustering problem is a focus in the study of the current clustering analysis.An improved high- dimensional data clustering algorithm based on super- network is proposed.Firstly,the high- dimensional data is mapped to a mass weighted network;Secondly,the edge weights of super network is defined.Again,the optimization of the hypergraph partitioning method is used to divide the weighted network.Finally,the high- dimensional data clustering is realized.This method filters out noise in clustering data effectively,and avoids the traditional clustering methods in the process of dimension reduction of defects.Experiments show that the algorithm is effective and accuracy.更多还原