CSDGCN

This repository contains the demo code of the paper:

[Commonsense Knowledge Enhanced Sentiment Dependency Graph for Sarcasm Detection]

which has been accepted by IJCAI2023.

Dependencies

Please use COMET to generate commonsense of sentences in the dataset first, and then generate sentiment graphs and dependency graphs.
Train with command, optional arguments could be found in train.py

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
__pycache__		__pycache__
datasets		datasets
layers		layers
models		models
senticNet		senticNet
300_riloff_embedding_matrix.pkl		300_riloff_embedding_matrix.pkl
README.md		README.md
bucket_iterator.py		bucket_iterator.py
bucket_iterator.pyc		bucket_iterator.pyc
data_utils.py		data_utils.py
data_utils.pyc		data_utils.pyc
dependency_graph.py		dependency_graph.py
path.py		path.py
riloff_word2idx.pkl		riloff_word2idx.pkl
select_contrast.py		select_contrast.py
select_major.py		select_major.py
select_minor.py		select_minor.py
sentic.py		sentic.py
tackle_datasets.py		tackle_datasets.py
train.py		train.py