Hate Speech Detection in Portuguese with Naïve Bayes, SVM, MLP and Logistic Regression

Abstract

Even though social networks can provide free space for discussing ideas, people can also use them to propagate hate speech and, given the amount of written material in such networks, it becomes necessary to rely on automatic methods for identifying this problem. In this work, we set out to verify the use of some classic Machine Learning algorithms for the task of hate speech detection in tweets written in Portuguese, by testing four different models (SVM, MLP, Logistic Regression and Naïve Bayes) with different configurations. Results show that these algorithms produce better results (in terms of micro-averaged F1 score) than the LSTM used for benchmark, being also competitive to other results by the related literature

O Discurso de Ódio Homofóbico no Twitter a partir da Análise de Dados

“…Quatro modelos de aprendizado de máquina para classificac ¸ão foram estudados em Silva and Roman [2020]. Os autores avaliaram modelos gerados pelos algoritmos Naive Bayes, Support Vector Machine, Bayesian Logistic Regression e Multilayer Perceptron Neural Network (MLP), com diversas configurac ¸ões.…”

Desenvolvimento de um conjunto de dados com comentários extraídos da plataforma Twitch sobre o jogo League of Legends

“…Seguindo um mesmo direcionamento, os conceitos utilizados por [Nascimento et al 2019, Coutinho and Malheiros 2020, Silva and Roman 2020 definem como uma manifestação hostil, preconceituosa e que visa denegrir pessoas ou grupos baseados em características, como o gênero, religião, nacionalidade, deficiência, orientação sexual e condicionamento físico. Além disso, a manifestação deles ocorrem pela mídia tradicional (e.g.…”

Hate Speech Hashtag Classification on Twitter Using the Hybrid Classifier Method

“…In MLP, there are two important parameters, namely the function activation and optimization functions [11]. Studies shows that MLP had been used for hate speech detection [12] and cancer breast classification with 97,7% accuracy using logistic activation and Adam optimization function [13].…”