Misspelling Oblivious Word Embeddings

Code

• bitbucket.org/bedizel/moe
  OfficialIn papernone★ 0
• github.com/facebookresearch/moe
  none★ 0
• github.com/dleemiller/string-noise
  none★ 0

Abstract

In this paper we present a method to learn word embeddings that are resilient to misspellings. Existing word embeddings have limited applicability to malformed texts, which contain a non-negligible amount of out-of-vocabulary words. We propose a method combining FastText with subwords and a supervised task of learning misspelling patterns. In our method, misspellings of each word are embedded close to their correct variants. We train these embeddings on a new dataset we are releasing publicly. Finally, we experimentally show the advantages of this approach on both intrinsic and extrinsic NLP tasks using public test sets.

Tasks

Reproductions