Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2303.08601 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910503041761280 |
|---|---|
| author | Wang, Yequan Zhang, Hengran Sun, Aixin Meng, Xuying |
| author_facet | Wang, Yequan Zhang, Hengran Sun, Aixin Meng, Xuying |
| contents | Given comparative text, comparative relation extraction aims to extract two targets (\eg two cameras) in comparison and the aspect they are compared for (\eg image quality). The extracted comparative relations form the basis of further opinion analysis.Existing solutions formulate this task as a sequence labeling task, to extract targets and aspects. However, they cannot directly extract comparative relation(s) from text. In this paper, we show that comparative relations can be directly extracted with high accuracy, by generative model. Based on GPT-2, we propose a Generation-based Comparative Relation Extractor (GCRE-GPT). Experiment results show that \modelname achieves state-of-the-art accuracy on two datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2303_08601 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | GCRE-GPT: A Generative Model for Comparative Relation Extraction Wang, Yequan Zhang, Hengran Sun, Aixin Meng, Xuying Computation and Language Given comparative text, comparative relation extraction aims to extract two targets (\eg two cameras) in comparison and the aspect they are compared for (\eg image quality). The extracted comparative relations form the basis of further opinion analysis.Existing solutions formulate this task as a sequence labeling task, to extract targets and aspects. However, they cannot directly extract comparative relation(s) from text. In this paper, we show that comparative relations can be directly extracted with high accuracy, by generative model. Based on GPT-2, we propose a Generation-based Comparative Relation Extractor (GCRE-GPT). Experiment results show that \modelname achieves state-of-the-art accuracy on two datasets. |
| title | GCRE-GPT: A Generative Model for Comparative Relation Extraction |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2303.08601 |