Semantic Compositionality through Recursive Matrix-Vector Spaces.
Richard SocherBrody HuvalChristopher D. ManningAndrew Y. NgPublished in: EMNLP-CoNLL (2012)
Keyphrases
- vector space
- lie group
- riemannian manifolds
- graph classification
- intrinsic dimensionality
- symmetric positive definite
- similarity search
- euclidean space
- latent semantic indexing
- distance measure
- low dimensional
- feature vectors
- square root
- metric space
- positive definite
- pattern recognition
- covariance matrices
- geometric structure
- semantic similarity
- valued data
- covariance matrix
- feature selection