KaGRMN-DSG_ABSA

This repository contains the PyTorch source Code for our paper: Understand me, if you refer to Aspect Knowledge: Knowledge-aware Gated Recurrent Memory Network. Bowen Xing and Ivor Tsang.

Accepted by IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI).

Architectures

Our framework:

Details of KaGRMN cell:

Requirements

Our code relies on Python 3.6 and following libraries:

• transformers==1.1.0
• torch==1.2.0
• numpy==1.16.0
• tqdm==4.50.2
• scikit_learn==0.23.2

Run

    # Restaurant14
    python run.py --cuda_id 0 --dataset_name rest --seed 2021  --embedding_type bert --self_num_heads 3 --rel_num_heads 4 --dropout 0.3 --bert_lr 5e-5 --learning_rate 1e-5 --weight_decay 0.05 --stack_num 4 --n_gcn 2
  
    # Laptop14
    python run.py --cuda_id 0 --dataset_name laptop --seed 2021  --embedding_type bert --self_num_heads 3 --rel_num_heads 2 --dropout 0.3 --bert_lr 1e-5 --learning_rate 5e-5 --weight_decay 0.001 --stack_num 4 --n_gcn 2

    # Restaurant15
    python run.py --cuda_id 0 --dataset_name res15 --seed 2021 --embedding_type bert --self_num_heads 6 --rel_num_heads 6 --dropout 0.3 --bert_lr 3e-5 --learning_rate 2e-5 --weight_decay 0.05 --stack_num 2 --n_gcn 2 --logging_steps 25

Citation

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

@ARTICLE{9735156,  author={Xing, Bowen and Tsang, Ivor W.},  
journal={IEEE Transactions on Emerging Topics in Computational Intelligence},   
title={Understand Me, if You Refer to Aspect Knowledge: Knowledge-Aware Gated Recurrent Memory Network},  
year={2022},  
volume={},  
number={},  
pages={1-11},  
doi={10.1109/TETCI.2022.3156989}}