BGCA

This repo contains the data and code for our paper "Bidirectional Generative Framework for Cross-domain Aspect-based Sentiment Analysis" (BGCA) in ACL 2023.

Description

Specifically, our framework trains a generative model in both text-to-label and label-to-text directions. The former transforms each task into a unified format to learn domain-agnostic features, and the latter generates natural sentences from noisy labels for data augmentation, with which a more accurate model can be trained.

Requirements

This repo is developed using the following packages:

• transformers==4.18.0
• sentencepiece==0.1.96
• pytorch_lightning==0.8.1
• editdistance==0.6.0
• scikit-learn==0.24.2
• numpy==1.22.3
• tqdm==4.64.0

Usage

We conduct experiments on four ABSA tasks:

1. ATE

cd code
bash ../scripts/run_ate.sh

2. UABSA

cd code
bash ../scripts/run_uasa.sh

3. AOPE

cd code
bash ../scripts/run_aope.sh

4. ASTE

cd code
bash ../scripts/run_aste.sh

Code Sturcture

• constants.py Contains constant variables
• data_utils.py Contains code to prepare input & output for generative model
• eval_utils.py Contains code to extract sentiment elements and calculate metric
• main.py Contains code for the main function
• model_utils.py Contains code for model initialization
• preprocess.py Contains code to preprocess different task's data (ATE and UABSA share the same data)
• run_utils.py Contains code for training, where data_gene() is the key method for this repo.
• setup.py Contains code for setup such as args parsing

Note

1. Extract model in the code refers to text-to-label stage in the paper, and Gene model refers to label-to-text stage.

Citation

If the code is used in your research, please star our repo and cite our paper as follows:

@inproceedings{deng-etal-2023-bidirectional,
    title = "Bidirectional Generative Framework for Cross-domain Aspect-based Sentiment Analysis",
    author = "Deng, Yue  and
      Zhang, Wenxuan  and
      Pan, Sinno Jialin  and
      Bing, Lidong",
    booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2023",
    address = "Toronto, Canada",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.acl-long.686",
    pages = "12272--12285",
}