采用邻域关联性的非监督流形对齐算法

假设对于两个流形上关联性较强的样本点,其邻域点之间也会具有较强的关联性.基于此假设,提出一种新的非监督流形对齐算法,通过学习局部邻域之间的关联性,挖掘不同流形样本点间的关联性;然后,将两个流形样本点投影到共同的低维空间,同时保持所挖掘的关联性.结果表明:与传统的非监督流形对齐算法比较,文中算法能更准确地找出不同流形数据在低维空间的匹配点.

Unsupervised Manifold Alignment Algorithm Using Neighborhood Correlation

This paper propose a basic assumption: for the points sampled from two manifolds which have strong correlations, their neighbors also have stronger correlations. Based on this assumption, this paper propose a new unsupervised manifold alignment algorithm which using the local neighborhood correlation to construct the relationship between the data sample points from different manifolds, and then projecting two manifold data to a common low-dimensional space while preserve the discovering of the correlation. The numerical experiments show that compared with the traditional unsupervised manifold alignment algorithms, this proposed algorithm can find the matching points of different manifold data in the low-dimensional space more accurately.