Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2310.14505 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914708650459136 |
|---|---|
| author | Meng, Fanfei Wang, Chen-Ao |
| author_facet | Meng, Fanfei Wang, Chen-Ao |
| contents | We propose a novel framework based on the attention mechanism to identify the sentiment of a movie review document. Previous efforts on deep neural networks with attention mechanisms focus on encoder and decoder with fixed numbers of multi-head attention. Therefore, we need a mechanism to stop the attention process automatically if no more useful information can be read from the memory.In this paper, we propose an adaptive multi-head attention architecture (AdaptAttn) which varies the number of attention heads based on length of sentences. AdaptAttn has a data preprocessing step where each document is classified into any one of the three bins small, medium or large based on length of the sentence. The document classified as small goes through two heads in each layer, the medium group passes four heads and the large group is processed by eight heads. We examine the merit of our model on the Stanford large movie review dataset. The experimental results show that the F1 score from our model is on par with the baseline model. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_14505 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Sentiment analysis with adaptive multi-head attention in Transformer Meng, Fanfei Wang, Chen-Ao Computation and Language We propose a novel framework based on the attention mechanism to identify the sentiment of a movie review document. Previous efforts on deep neural networks with attention mechanisms focus on encoder and decoder with fixed numbers of multi-head attention. Therefore, we need a mechanism to stop the attention process automatically if no more useful information can be read from the memory.In this paper, we propose an adaptive multi-head attention architecture (AdaptAttn) which varies the number of attention heads based on length of sentences. AdaptAttn has a data preprocessing step where each document is classified into any one of the three bins small, medium or large based on length of the sentence. The document classified as small goes through two heads in each layer, the medium group passes four heads and the large group is processed by eight heads. We examine the merit of our model on the Stanford large movie review dataset. The experimental results show that the F1 score from our model is on par with the baseline model. |
| title | Sentiment analysis with adaptive multi-head attention in Transformer |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2310.14505 |