Misspelling Oblivious Word Embeddings

Bora Edizel, Aleksandra Piktus, Piotr Bojanowski, Rui Ferreira, Edouard Grave +1 more
5/23/2019
cs.CLcs.LG

No official code found. Best implementation: facebookresearch/moe (201 stars).

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.

View on arXivView PDF

Code Implementations(1)

Misspelling Oblivious Word Embeddings

20122

Similar Papers

Cite this paper

@article{edizel2019misspelling,
  title  = {Misspelling Oblivious Word Embeddings},
  author = {Bora Edizel and Aleksandra Piktus and Piotr Bojanowski and Rui Ferreira and Edouard Grave and Fabrizio Silvestri},
  year   = {2019},
  eprint = {1905.09755},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1905.09755}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1905.09755.svg)](https://codeofpaper.com/paper/1905.09755)

Discussion