Few-Shot Transfer Learning for Text Classification With Lightweight Word Embedding Based Models
| Autor: | Gong Jianxing, Jian Huang, Chongyu Pan, Xingsheng Yuan |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
text classification
Word embedding General Computer Science Computer science Few-shot learning Feature extraction Pooling 02 engineering and technology transfer learning 010501 environmental sciences computer.software_genre 01 natural sciences Data modeling pooling strategy 0202 electrical engineering electronic engineering information engineering General Materials Science 0105 earth and related environmental sciences word embedding based models business.industry Deep learning 020208 electrical & electronic engineering General Engineering Problem domain Task analysis lcsh:Electrical engineering. Electronics. Nuclear engineering Artificial intelligence Transfer of learning business lcsh:TK1-9971 computer Word (computer architecture) Natural language processing |
| Zdroj: | IEEE Access, Vol 7, Pp 53296-53304 (2019) |
| ISSN: | 2169-3536 |
| DOI: | 10.1109/access.2019.2911850 |
| Popis: | Many deep learning architectures have been employed to model the semantic compositionality for text sequences, requiring a huge amount of supervised data for parameters training, making it unfeasible in situations where numerous annotated samples are not available or even do not exist. Different from data-hungry deep models, lightweight word embedding-based models could represent text sequences in a plug-and-play way due to their parameter-free property. In this paper, a modified hierarchical pooling strategy over pre-trained word embeddings is proposed for text classification in a few-shot transfer learning way. The model leverages and transfers knowledge obtained from some source domains to recognize and classify the unseen text sequences with just a handful of support examples in the target problem domain. The extensive experiments on five datasets including both English and Chinese text demonstrate that the simple word embedding-based models (SWEMs) with parameter-free pooling operations are able to abstract and represent the semantic text. The proposed modified hierarchical pooling method exhibits significant classification performance in the few-shot transfer learning tasks compared with other alternative methods. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|