一种基于海量高维数据的软子空间聚类改进算法

摘要

基于加权的软子空间聚类是处理高维数据的一种有效手段,在原有的软子空间聚类算法基础上对目标函数进行改进,从而结出一种新的软子空间聚类算法.相较之原算法,此算法具有更高的抗噪性及聚类效率;通过与典型的软子空间聚类算法比较试验,结果表明可有效提高海量高维数据的聚类处理效率.%The weighted soft subspace clustering method is an effective tool to process high-dimensional data. In this paper a new subspace clustering algorithm is proposed by improving the objective function of the original sub-space clustering algorithm, and the new algorithm has a good anti -noised performance. Compared with several typical soft subspace clustering algorithms,the experiment results show that the efficiency of mass high-dimension-al data clustering is fairly improved.