Local Neighbor Propagation Embedding

Manifold learning occupies a vital role in the field of nonlinear dimensionality reduction and serves many relevant methods. Typical linear embedding methods have achieved remarkable results. However, these methods perform poorly on sparse data. To address this issue, we introduce neighbor propagation into Locally Linear Embedding (LLE) and propose a new method named Local Neighbor Propagation Embedding (LNPE). LNPE enhances the local connections between neighborhoods by extending 1-hop neighbors into n-hop neighbors. The experimental results show that LNPE can obtain faithful embeddings with better topological and geometrical properties.