An unsupervised lemmatization method that clusters word forms with a distance measure combining Jaro-Winkler edit distance and FastText embedding cosine similarity surpasses simple baselines on most of 28 Universal Dependencies treebanks.
Dryer and Martin Haspelmath, editors
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Unsupervised Lemmatization as Embeddings-Based Word Clustering
An unsupervised lemmatization method that clusters word forms with a distance measure combining Jaro-Winkler edit distance and FastText embedding cosine similarity surpasses simple baselines on most of 28 Universal Dependencies treebanks.