Detecção de discurso de ódio em português usando CNN combinada a vetores de palavras

• Samuel C. Silva UNESP
• Adriane B. S. Serapião UNESP

DOI: https://doi.org/10.5753/kdmile.2018.27378

Resumo

The current work has proposed to study and to implement a convolutional neural network (CNN) allied to pre-trained (Wang2Vec and GloVe) and trainable word embeddings for hate speech detection in Portuguese. For sake of comparison, the implementation used different gradient descent optimizer functions (RMSprop, Adagrad, Adadelta and Adam), aiming to contrast the performance at each function. For such task, it were used three datasets of comments in Portuguese, annotated as offensive or not offensive. We have concluded that using this proposed approach the results were superior to those from the baseline, achieving higher F-score and accuracy measures.

Palavras-chave: convolutional neural networks, hate speech, natural language processing

Referências

Aggarwal, C. C. Data Mining: The Textbook. Springer Publishing Company, Incorporated, 2015.

Almeida, T. G., Nakamura, F. G., and Nakamura, E. F. Uma abordagem para identificar e monitorar haters em redes sociais online, 2017.

Cohen-Almagor, R. Fighting hate and bigotry on the internet. Policy & Internet 3 (3): 1–26, 2011.

Fortuna, P. C. T. Automatic detection of hate speech in text: an overview of the topic and dataset annotation with hierarchical classes, 2017.

Georgakopoulos, S. V., Tasoulis, S. K., Vrahatis, A. G., and Plagianakos, V. P. Convolutional neural networks for toxic comment classification. arXiv preprint arXiv:1802.09957, 2018.

Hartmann, N., Fonseca, E., Shulby, C., Treviso, M., Rodrigues, J., and Aluisio, S. Portuguese word embeddings: Evaluating on word analogies and natural language tasks. arXiv preprint arXiv:1708.06025, 2017.

Kim, Y. Convolutional neural networks for sentence classification. arXiv preprint arXiv:1408.5882, 2014.

LeCun, Y., Bengio, Y., and Hinton, G. Deep learning. Nature 521 (7553): 436–444, 5, 2015.

Lopez, M. M. and Kalita, J. Deep learning applied to nlp. CoRR vol. abs/1703.03091, 2017.

Malmasi, S. and Zampieri, M. Detecting hate speech in social media. arXiv preprint arXiv:1712.06427, 2017.

Nobata, C., Tetreault, J., Thomas, A., Mehdad, Y., and Chang, Y. Abusive language detection in online user content. In Proceedings of the 25th international conference on world wide web. International World Wide Web Conferences Steering Committee, pp. 145–153, 2016.

Pelle, R. and Moreira, V. Offensive comments in the brazilian web: a dataset and baselines results. In Proc. of the 6th Brazilian Workshop on Social Network Analysis and Mining. pp. 1–160, 2017.

Pitsilis, G. K., Ramampiaro, H., and Langseth, H. Detecting offensive language in tweets using deep learning. arXiv preprint arXiv:1801.04433, 2018.

Ruder, S. An overview of gradient descent optimisation algorithms. arXiv preprint arXiv:1609.04747, 2016.

Schmidt, A. and Wiegand, M. A survey on hate speech detection using natural language processing. In Proceedings of the Fifth International Workshop on Natural Language Processing for Social Media. pp. 1–10, 2017.

Yin, W., Kann, K., Yu, M., and Schütze, H. Comparative study of cnn and rnn for natural language processing. arXiv preprint arXiv:1702.01923, 2017.

Zhang, Y. and Wallace, B. A sensitivity analysis of (and practitioners’ guide to) convolutional neural networks for sentence classification. arXiv preprint arXiv:1510.03820, 2015.

Zhang, Z., Robinson, D., and Tepper, J. Detecting hate speech on twitter using a convolution-gru based deep neural network. In European Semantic Web Conference. Springer, pp. 745–760, 2018.