Knowledge Graph

Graph explorer

Search resolved entities, walk their edges, and read the sentence behind every one.

Entities25 of 207,987

Entity

t-SNE

technology4 mentionslast seen Nov 9, 2026

A neighbor embedding method that prioritizes local similarity preservation in dimensionality reduction.

Relationships 1

t-SNEcompetes withLaplacian Eigenmaps85% confidence1 source

Relationship

t-SNEcompetes withLaplacian Eigenmaps

85% confidence1 supporting mentions

Either name opens that entity’s relationships.

Sources 1

preprintarXivSep 15, 2026A Spectral Decomposition Framework for Multiscale Nonlinear Dimensionality Reduction
Neighbor embedding methods such as t-SNE and UMAP prioritize local similarity preservation but do not explicitly constrain global organization, whereas standard spectral methods such as Laplacian Eigenmaps capture smooth, coarse-scale graph structure