系数增强最小二乘回归子空间聚类法  

Coefficient Enhanced Least Square Regression Subspace Clustering Method

作　　者：简彩仁[1] 翁谦[2] 夏靖波[1] JIAN Cairen;WENG Qian;XIA Jingbo(School of Information Science&Technology,Tan Kah Kee Colleage,Xiamen University,Zhangzhou,Fujian 363105,China;College of Mathematics and Computer Science,Fuzhou University,Fuzhou 350108,China)

机构地区：[1]厦门大学嘉庚学院信息科学与技术学院,福建漳州363105 [2]福州大学数学与计算机科学学院,福州350108

出　　处：《计算机工程与应用》2022年第20期73-78,共6页Computer Engineering and Applications

基　　金：国家自然科学基金(41801324);福建省自然科学基金(2019J01244,2018J07005)。

摘　　要：针对最小二乘回归子空间聚类法在求解表示系数时忽略了样本相似度的不足,提出改进方法。基于样本相互重构的表示系数矩阵和样本相似度矩阵有很大的关联定义系数增强项,求解可以保持样本相似度的表示系数矩阵,提出系数增强最小二乘回归子空间聚类法。在8个标准数据集上的实验表明该方法可以提高最小二乘回归子空间聚类法的聚类性能。In view of the fact that the least square regression subspace clustering method ignores the similarity between samples when solving the representation coefficients,an improved method is proposed.Based on the representation coeffi-cient matrix of the sample mutual reconstruction and the similarity matrix of the sample having a great correlation,it defines the coefficient enhancement term to solve the representation coefficient matrix that can preserve sample similarity.Experi-ments on 8 standard data sets show that the proposed method can improve the performance of least square regression sub-space clustering method.

关 键 词：最小二乘回归 子空间聚类 系数增强 高维数据 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论] TP371[自动化与计算机技术—计算机科学与技术]