NMF variant learns local similarity and clustering mutually to yield representations that better reveal data geometry, with nonlinear extension and convergent multiplicative updates.
On the equivalence of nonnegative matrix factorization and spectral clustering,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Nonnegative Matrix Factorization with Local Similarity Learning
NMF variant learns local similarity and clustering mutually to yield representations that better reveal data geometry, with nonlinear extension and convergent multiplicative updates.