中文摘要：

局部线性判别嵌入（locally linear discriminant embedding,LLDE）将局部线性嵌入（locally linear embedding,LLE）和最大间隔（maximum margin criterion,MMC）进行融合,有效地提高了LLE算法的识别力。但其保留的是数据的全局判别信息,且依赖数据的分布。针对LLDE的不足,本研究将LLE和加权非参数最大间隔（weighted non-parametric maximum margin criterion,WNMMC）进行融合,提出了一种新的有监督的降维方法——非参数判别性局部线性嵌入（nonparametric locally linear discriminant embedding,NLLDE）。NLLDE保留了数据更为有效的局部判别信息,因此更具判别力。NLLDE采用了非参数数据表示,使得模型及求解不依赖于数据的分布,克服了LLDE针对高斯分布数据有效的局限,其应用范围更为广泛。Yale和PIE人脸数据库上的实验结果证实了NLLDE的高效性。

英文摘要：

Locally linear discriminant embedding（LLDE） can effectively enhance the discriminability of locally linear embedding（LLE） by adding the maximum margin criterion（MMC） into the objective function of LLE.However,LLDE seeks to preserve the global discriminative information of the sample and the optimal result is only achieved when the data is of Gaussian distribution.A novel supervised dimensionality reduction method,namely nonparametric locally linear discriminant embedding（NLLDE）,was proposed by adding the weighted nonparametric maximum margin criterion（WNMMC） into the objective function of LLE to overcome the drawbacks of LLDE.NLLDE explored the local discriminative information of the data,which had more discriminating power.Furthermore,NLLDE did not assume the particular form of class densities.This meant that NLDE could be applied in more fields.The experimental results on Yale and PIE face database indicated the effectiveness of this method.