This repository provides code for the paper "Hate Speech Detection based on Sentiment Knowledge Sharing" with adjustments made by Aryan Chawla and Ronald Sun.
They are illustrated in the requirements.txt file. However, it is reccomended to install the dependencies one-by-one rather then with -r option. If pyenchant errors occur, please use pyenchant==3.1.0. Finally, you may need to interface with python and install punkt if it is the first time using nltk and/or the project.
The SE dataset may need some adjustment in formatting from tsv to csv. Make sure to put these in the data directory and also within their respective directory too. ex: SemEval_task5/df_test.csv
The glove txt file can be downloaded here. There is also a larger one available, but make sure to adjust the script for it here
After download the data and the pre-trained word vectors, just run the bash script associated with the datset.