25-d GloVe Model Pretrained on Twitter (subset)

An embeddings object containing a subset of glove.twitter.25d pretrained by Pennington et al. (2014) on 2B Tweets. This subset includes only embeddings included in the 12,000 most common English words, as defined by Peter Norvig's compilation, derived from the Google Web Trillion Word Corpus. Full pretrained GloVe models are available for download at https://nlp.stanford.edu/projects/glove/

Usage

glove_twitter_25d

Format

An object of class embeddings (inherits from matrix, array) with 11925 rows and 25 columns.

References

Pennington, J., Socher, R., and Manning, C. D. (2014). GloVe: Global Vectors for Word Representation. https://nlp.stanford.edu/projects/glove/