摘要：Locally linear embedding (LLE) is an efficient dimensional reduction algorithm for nonlinear data， and the low dimensional data can maintain topological relations in the original space after the processing. But this algorithm main application is not very good in the data dimensional reduction, the visualization and learning effects of data classification question and so on. In ordered to solve the above question, this paper proposes an efficient dimensional reduction and data classification method--local graph embedding method based on maximum margin criterion (LGE/MMC) for dimensional reduction, which is applied in face recognition. This goal of algorithm is preserved under nearest neighbour premise, where MMC criterion is used to construct the intrinsic graph and the penalty graph. In the intrinsic graph, the nonlinear structure is discovered in the high dimensional data space by the locally symmetric of linear restructuring, which is caused the similar sample as far as possible to gather in together. At the same time, the different class sample is far away as far as possible in the penalty graph. LGE/MMC seeks to minimize the difference, rather than the ratio, between the locality preserving between-class scatter and locality preserving within-class scatter. The results of face recognition experiments on ORL, YALE and AR face databases demonstrate the effectivity of the proposed method.
关键词：locally linear embedding; dimensional reduction; face recognition; maximum margin criterion; local graph;embedding
