Using Convolutional Neural Networks to Classify Hate-Speech

Björn Gambäck, Utpal Kumar Sikdar


Abstract
The paper introduces a deep learning-based Twitter hate-speech text classification system. The classifier assigns each tweet to one of four predefined categories: racism, sexism, both (racism and sexism) and non-hate-speech. Four Convolutional Neural Network models were trained on resp. character 4-grams, word vectors based on semantic information built using word2vec, randomly generated word vectors, and word vectors combined with character n-grams. The feature set was down-sized in the networks by max-pooling, and a softmax function used to classify tweets. Tested by 10-fold cross-validation, the model based on word2vec embeddings performed best, with higher precision than recall, and a 78.3% F-score.
Anthology ID:
W17-3013
Volume:
Proceedings of the First Workshop on Abusive Language Online
Month:
August
Year:
2017
Address:
Vancouver, BC, Canada
Editors:
Zeerak Waseem, Wendy Hui Kyong Chung, Dirk Hovy, Joel Tetreault
Venue:
ALW
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
85–90
Language:
URL:
https://aclanthology.org/W17-3013
DOI:
10.18653/v1/W17-3013
Bibkey:
Cite (ACL):
Björn Gambäck and Utpal Kumar Sikdar. 2017. Using Convolutional Neural Networks to Classify Hate-Speech. In Proceedings of the First Workshop on Abusive Language Online, pages 85–90, Vancouver, BC, Canada. Association for Computational Linguistics.
Cite (Informal):
Using Convolutional Neural Networks to Classify Hate-Speech (Gambäck & Sikdar, ALW 2017)
Copy Citation:
PDF:
https://aclanthology.org/W17-3013.pdf