福州大学学报(自然科学版)
福州大學學報(自然科學版)
복주대학학보(자연과학판)
JOURNAL OF FUZHOU UNIVERSITY(NATURAL SCIENCE EDITION)
2015年
4期
466-470
,共5页
降维%近邻保持嵌入%判别信息%局部结构
降維%近鄰保持嵌入%判彆信息%跼部結構
강유%근린보지감입%판별신식%국부결구
dimensionality reduction%neighborhood preserving embedding%discrimination informa-tion%local structure
针对传统近邻保持嵌入算法( NPE)侧重保持样本的局部结构,而没有考虑样本类别信息的不足,提出判别局部近邻保持嵌入算法DLNPE.该算法利用样本点的局部结构构造新定义下的类内类间散布矩阵,并以此作为判别信息引入目标函数.在6个真实数据上进行实验,证明了所提算法的有效性.
針對傳統近鄰保持嵌入算法( NPE)側重保持樣本的跼部結構,而沒有攷慮樣本類彆信息的不足,提齣判彆跼部近鄰保持嵌入算法DLNPE.該算法利用樣本點的跼部結構構造新定義下的類內類間散佈矩陣,併以此作為判彆信息引入目標函數.在6箇真實數據上進行實驗,證明瞭所提算法的有效性.
침대전통근린보지감입산법( NPE)측중보지양본적국부결구,이몰유고필양본유별신식적불족,제출판별국부근린보지감입산법DLNPE.해산법이용양본점적국부결구구조신정의하적류내류간산포구진,병이차작위판별신식인입목표함수.재6개진실수거상진행실험,증명료소제산법적유효성.
Because traditional neighborhood preserving embedding ( NPE) focuses on keeping a sam-ple of local structure without taking category information into account, the discriminant local neighbor-hood preserving embedding algorithm is proposed.The algorithm use the local structure of samples points to construct the new definition within-class and between-class scatter matrix, which are intro-duced into the objective function as the discrimination information.The experimental results on six real data show that the algorithm is effective.