We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Tweet2Vec: Character-Based Distributed Representations for Social Media

Abstract: Text from social media provides a set of challenges that can cause traditional NLP approaches to fail. Informal language, spelling errors, abbreviations, and special characters are all commonplace in these posts, leading to a prohibitively large vocabulary size for word-level approaches. We propose a character composition model, tweet2vec, which finds vector-space representations of whole tweets by learning complex, non-local dependencies in character sequences. The proposed model outperforms a word-level baseline at predicting user-annotated hashtags associated with the posts, doing significantly better when the input contains many out-of-vocabulary words or unusual character sequences. Our tweet2vec encoder is publicly available.
Comments: 6 pages, 2 figures, 4 tables, accepted as conference paper at ACL 2016
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Cite as: arXiv:1605.03481 [cs.LG]
  (or arXiv:1605.03481v2 [cs.LG] for this version)

Submission history

From: Bhuwan Dhingra [view email]
[v1] Wed, 11 May 2016 15:30:09 GMT (559kb,D)
[v2] Tue, 17 May 2016 15:00:38 GMT (559kb,D)

Link back to: arXiv, form interface, contact.