Skip to content

Latest commit

 

History

History
35 lines (26 loc) · 1.49 KB

README.md

File metadata and controls

35 lines (26 loc) · 1.49 KB
Raw

DREGCN

Code for A Dependency Syntactic Knowledge Augmented Interactive Architecture for End-to-End Aspect-based Sentiment Analysis

Introduction

The implementation is based on IMN and the dataset we used is also from IMN. The dependency tree is generated by spaCy. Download Glove file and the bert-based feature used in this Repo is produced by BERT. The dependency tree (i.e., *.pkl) is generated by prepare_data_A.py and utils_opinion.py, most of the two files are adapted on this.

Usage

Training and evaluating with the following scripts (the Hyper-parameters are shown in 'train.py' file and you may change them for better results):

bash train.sh

Requirements

  • Python 2.7
  • Keras 2.2.4
  • tensorflow 1.4.1

Citation

If you find this project helps, please cite the following paper :)

@article{LIANG2021,
title = {A Dependency Syntactic Knowledge Augmented Interactive Architecture for End-to-End Aspect-based Sentiment Analysis},
journal = {Neurocomputing},
year = {2021},
issn = {0925-2312},
doi = {https://doi.org/10.1016/j.neucom.2021.05.028},
url = {https://www.sciencedirect.com/science/article/pii/S0925231221007815},
author = {Yunlong Liang and Fandong Meng and Jinchao Zhang and Yufeng Chen and Jinan Xu and Jie Zhou}
}