AWhat changed
Relates random-walk/negative-sampling embedding objectives to graph matrix factorizations and computes clipped log co-occurrence embeddings directly or with spectral approximation.
Scientific scope
Relates random-walk/negative-sampling embedding objectives to graph matrix factorizations and computes clipped log co-occurrence embeddings directly or with spectral approximation. Original derivation assumes connected undirected non-bipartite graphs and stationary starts or a long-walk limit, together with inherited idealized SGNS factorization. Finite samples, low-rank embeddings and nonconvex training are not literally identical to the population optimum. NetMF truncates/clips its matrix and approximates large windows; general method only, bio application results omitted.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Original derivation assumes connected undirected non-bipartite graphs and stationary starts or a long-walk limit, together with inherited idealized SGNS factorization. Finite samples, low-rank embeddings and nonconvex training are not literally identical to the population optimum. NetMF truncates/clips its matrix and approximates large windows; general method only, bio application results omitted.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction