基于加权集成Nystr?m采样的谱聚类算法  

Spectral Clustering Algorithm Based on Weighted Ensemble Nystr?m Sampling

作　　者：邱云飞[1] 刘畅[1] QIU Yunfei;LIU Chang(School of Software,Liaoning Technical University,Huludao 125105)

机构地区：[1]辽宁工程技术大学

出　　处：《模式识别与人工智能》2019年第5期420-428,共9页Pattern Recognition and Artificial Intelligence

基　　金：国家自然科学基金项目(No.71771111)资助~~

摘　　要：针对Nystrom方法在谱聚类应用中存在聚类效果不稳定、样本代表性较弱的问题,提出基于加权集成Nystrom采样的谱聚类算法.首先利用统计杠杆分数区别数据间的重要程度,对数据进行加权.然后基于权重采用加权K-means中心点采样,得到多组采样点.再引入集成框架,利用集群并行运行Nystrom方法构建近似核矩阵.最后利用岭回归方法组合各个近似核矩阵,产生比标准Nystrom方法更准确的低秩近似.在UCI数据集上的测试实验表明,文中算法取得较理想的聚类结果.Since most Nystrom methods have problems of unstable clustering effect and weak representativeness in spectral clustering application,a spectral clustering algorithm based on weighted ensemble Nystrom sampling is proposed.Firstly,the statistical leverage score is used to distinguish the importance of data and the data are weighted.Then,based on these weights,the weighted K-means center point sampling is used to obtain multiple sets of sampling points.The integration framework is introduced,and the approximate kernel matrix is constructed using the cluster parallel operation Nystrom method.Finally,the approximate kernel is determined by the ridge regression method.The matrices are combined to produce a more accurate low rank approximation than that by standard Nystrom method.Experiments on UCI datasets demonstrate that the proposed algorithm achieves better clustering results.

关 键 词：谱聚类 Nystrom采样 统计杠杆分数加权 集成Nystrom 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]