Domain Adaptation for Text Classification with Weird Embeddings
Pre-trained word embeddings are often used to initialize deep learning models for text classification, as a way to inject precomputed lexical knowledge and boost the learning process. However, such embeddings are usually trained on generic corpora, while text classification tasks are often domain-sp...
Gardado en:
| Autor Principal: | |
|---|---|
| Formato: | Chapter |
| Idioma: | Inglês |
| Publicado: |
Accademia University Press
2020
|
| Acceso en liña: | https://doi.org/10.4000/books.aaccademia.8250 https://hdl.handle.net/20.500.13089/1dki https://books.openedition.org/aaccademia/8250 |
| Tags: |
Sen Etiquetas, Sexa o primeiro en etiquetar este rexistro!
|