摘要
为了解决核主分量分析方法处理大训练样本集时计算代价巨大的问题,在采用子集划分的KPCA算法基础上,提出采用核聚类划分子集,并用每个子集的协方差矩阵的特征值累积贡献率作为标准来选取相应的特征向量.分别在人工和实际数据集上测试,实验结果显示在同一累积贡献率和给定子集个数的条件下,采用核聚类划分子集总能得到较小尺寸的核矩阵,而核矩阵尺寸的减小有助于改善测试样本的特征提取速度以及降低特征分解核矩阵的时间复杂度.
To overcome the computational problems of the standard kernel principal component analysis （ KPCA） algorithm,the authors proposed a new method for eigenvector selection by evaluating the cumulative contribution rate of the eigenvalues of the covariance matrix.In addition,a new way to partition the training data set based on kernel clustering was also developed.The influence was then explored of different partitions of training data sets on the size of the final kernel matrix,on the conditions causing a given cumulative contribution rate,and on the num-ber of subsets.Experimental results showed that a smaller kernel matrix can be obtained when kernel clustering method are used to partition the training dataset.The proposed algorithm can be helpful to reduce the time complex-ity of the eigen decomposition of a kernel matrix and to improve the speed of feature extraction for test samples.
出处
《智能系统学报》
2010年第3期221-226,共6页
CAAI Transactions on Intelligent Systems
基金
国家"863"计划资助项目(2007AA04Z423
2006AA01Z106)
国家自然科学基金资助项目(60576033)
福建省自然科学基金资助项目(2008J04001)
厦门市科技计划资助项目(3502Z20083031)
关键词
核主分量分析
核聚类
子集划分
协方差矩阵
特征向量
KPCA
kernel clustering
partition of training data set
covariance matrix
eigenvector
