#Sentic GCN

Introduction

This repository was used in our paper:

Aspect-Based Sentiment Analysis via Affective Knowledge Enhanced Graph Convolutional Networks
Bin Liang, Hang Su, Lin Gui, Erik Cambria, Ruifeng Xu. Knowledge-Based Systems, 2021: 107643.

Please cite our paper and kindly give a star for this repository if you use this code.

Requirements

• Python 3.6
• PyTorch 1.0.0
• SpaCy 2.0.18
• numpy 1.15.4

Usage

• Please download SenticNet from https://sentic.net/downloads/.
• Install SpaCy package and language models with

pip install spacy

and

python -m spacy download en

• Generate dependency graph with

python generate_dependency_graph.py

• Generate sentic graph with

python generate_sentic_graph.py

Training

• Train with command, optional arguments could be found in train.py & train_bert.py

• Run senticgcn: ./run_senticgcn.sh

• Run senticgcn_bert: ./run_senticgcn_bert.sh

Citation

The BibTex of the citation is as follow:

@article{liang2021aspect,
  title={Aspect-based sentiment analysis via affective knowledge enhanced graph convolutional networks},
  author={Liang, Bin and Su, Hang and Gui, Lin and Cambria, Erik and Xu, Ruifeng},
  journal={Knowledge-Based Systems},
  pages={107643},
  year={2021},
  publisher={Elsevier}
}

Credits

• The affective knowledge used in this work is from SenticNet, I would like to express my heartfelt thanks to all the authors of SenticNet.
• The code of this repository partly relies on ASGCN & ABSA-PyTorch.
• Here, I would like to express my gratitude to the authors of the ASGCN & ABSA-PyTorch repositories.