Neighbor Line-Based Locally Linear Embedding

@incollection{citeulike:2969521, abstract = {Locally linear embedding (Lle) is a powerful approach for mapping high-dimensional data nonlinearly to a lower-dimensional space. However, when the training examples are not densely sampled, Lle often returns invalid results. In this paper, the Nl3e (Neighbor Line-based Lle) approach is proposed, which generates some virtual examples with the help of neighbor line such that the Lle learning can be executed on an enriched training set. Experiments show that Nl3e outperforms Lle in visualization.}, author = {Zhan, De-Chuan and Zhou, Zhi-Hua }, citeulike-article-id = {2969521}, journal = {Advances in Knowledge Discovery and Data Mining}, keywords = {lle, manifold}, pages = {806--815}, posted-at = {2008-07-07 12:28:26}, priority = {2}, title = {Neighbor Line-Based Locally Linear Embedding}, url = {http://dx.doi.org/10.1007/11731139\_94}, year = {2006} }

TY - CHAP ID - citeulike:2969521 L3 - citeulike-article-id:2969521 TI - Neighbor Line-Based Locally Linear Embedding JF - Advances in Knowledge Discovery and Data Mining SP - 806 EP - 815 N2 - Locally linear embedding (Lle) is a powerful approach for mapping high-dimensional data nonlinearly to a lower-dimensional space. However, when the training examples are not densely sampled, Lle often returns invalid results. In this paper, the Nl3e (Neighbor Line-based Lle) approach is proposed, which generates some virtual examples with the help of neighbor line such that the Lle learning can be executed on an enriched training set. Experiments show that Nl3e outperforms Lle in visualization. KW - lle KW - manifold AU - Zhan, De-Chuan AU - Zhou, Zhi-Hua PY - 2006/// UR - http://dx.doi.org/10.1007/11731139_94 ER -