# Bojanowski 2017: Enriching Word Vectors with Subword Information

> Piotr Bojanowski, Edouard Grave, Armand Joulin et al., 2017. 1 records on Mysteries Department cite it. The inductive-methods survey proposes its subword…

Source: https://mysteriesdept.com/sources/SRC0D8MQ7F02GWR/enriching-word-vectors-with-subword-information  
Publisher: Mysteries Department  
Licence: CC BY 4.0 (credit Mysteries Department and link the page)  

---

[Sources](https://mysteriesdept.com/sources)[Voynich Manuscript](https://mysteriesdept.com/mysteries/voynich-manuscript)Bojanowski 2017

**0**

# Enriching Word Vectors with Subword Information

Piotr Bojanowski, Edouard Grave, Armand Joulin et al., 2017. _Transactions of the Association for Computational Linguistics_.

The fastText approach to word vectors: within a skip-gram model, a word's vector is the sum of vectors for the character n-grams it contains, so vectors can also be built for words never seen in training.

**Voynich Manuscript** Articledoi: **[10.1162/tacl_a_00051](https://doi.org/10.1162/tacl_a_00051)** Checked against the original0

## What it's used for here

The inductive-methods survey proposes its subword vectors as a spelling-only comparison when building embeddings from the small Voynich corpus.

## Cited by on this site 1 record

[1 methods surveys]

Inductive methods

SRC

## Cite it

```
Bojanowski, P., Grave, E., Joulin, A., & Mikolov, T. (2017). Enriching Word Vectors with Subword Information. Transactions of the Association for Computational Linguistics. https://doi.org/10.1162/tacl_a_00051
```

[Read it free at doi.org](https://doi.org/10.1162/tacl_a_00051)[PDF stored here · 12 pages](https://mysteriesdept.com/sources/SRC0D8MQ7F02GWR/enriching-word-vectors-with-subword-information/pdf)

The PDF is the copy doi.org gives away, kept here so the link can't break. Downloads are open to beta members.

Key

**SRC** 0D8MQ7F02GWR

Kind

Journal article

Rights

CC-BY-4.0

Checked

Against the published article

Added by

[apollo](https://mysteriesdept.com/u/apollo)

The record

Bibliographic records are CC0. Take this one as CSL-JSON, BibTeX or RIS.

[CSL-JSON](https://mysteriesdept.com/sources/SRC0D8MQ7F02GWR/enriching-word-vectors-with-subword-information/csl.json)