Unsupervised Linear Discriminant Analysis

作     者：唐宏 方涛 施鹏飞 唐国安 

作者机构：Inst.of Image Processing & Pattern RecognitionShanghai Jiaotong Univ Shanghai 200030 China Three Gorges Project Command of Armed Police Hydropower Eng.Troops Hubei 443133 

出 版 物：《Journal of Shanghai Jiaotong university(Science)》 (上海交通大学学报（英文版）)

年 卷 期：2006年第11卷第1期

页      面：40-42页

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 

基　　金：TheHighTechniqueProgramofChina(No.2001AA135091)andtheNationalNaturalScienceFoundationofChina(No.60275021) 

主　　题：linear discriminant analysis(LDA) unsupervised learning neighbor graph 

摘      要：An algorithm for unsupervised linear discriminant analysis was presented. Optimal unsupervised discriminant vectors are obtained through maximizing covariance of all samples and minimizing covariance of local k-nearest neighbor samples. The experimental results show our algorithm is effective.