Locally Linear Embedding Algorithm with Adaptive Neighbors

How to choose a proper number of the neighbors is an important issue of the locally linear embedding algorithm. To investigate this issue, we propose an optimized locally linear embedding algorithm with adaptive neighbors (ANLLE). The ANLLE selects the neighbors with a locally adaptive criterion. In addition, a new data point mapping method that computes the low-dimensional description of the correspondents is introduced in the ANLLE. The experiment results of the manifold expansion and the face recognition showed that the optimized algorithm is more effective than the original algorithm. The result of the present work implied that the ANLLE could improve the linear correlation of the neighbors and the data points. Moreover, it maintains the distance between the data points and reduces the application difficulty of the locally linear embedding algorithm.