Word Embeddings
fastText
Introduced by Bojanowski et al. in Enriching Word Vectors with Subword InformationfastText embeddings exploit subword information to construct word embeddings. Representations are learnt of character $n$-grams, and words represented as the sum of the $n$-gram vectors. This extends the word2vec type models with subword information. This helps the embeddings understand suffixes and prefixes. Once a word is represented using character $n$-grams, a skipgram model is trained to learn the embeddings.
Source: Enriching Word Vectors with Subword Information
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Text Classification | 36 | 7.44% |
| Sentence | 28 | 5.79% |
| General Classification | 28 | 5.79% |
| Sentiment Analysis | 24 | 4.96% |
| Classification | 16 | 3.31% |
| Named Entity Recognition (NER) | 15 | 3.10% |
| Language Modeling | 12 | 2.48% |
| Language Modelling | 12 | 2.48% |
| Word Similarity | 11 | 2.27% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
