Automatic Hate Speech Detection using Machine Learning: A Comparative Study

Abstract

The increasing use of social media and information sharing has given major benefits to humanity. However, this has also given rise to a variety of challenges including the spreading and sharing of hate speech messages. Thus, to solve this emerging issue in social media sites, recent studies employed a variety of feature engineering techniques and machine learning algorithms to automatically detect the hate speech messages on different datasets. However, to the best of our knowledge, there is no study to compare the variety of feature engineering techniques and machine learning algorithms to evaluate which feature engineering technique and machine learning algorithm outperform on a standard publicly available dataset. Hence, the aim of this paper is to compare the performance of three feature engineering techniques and eight machine learning algorithms to evaluate their performance on a publicly available dataset having three distinct classes. The experimental results showed that the bigram features when used with the support vector machine algorithm best performed with 79% off overall accuracy. Our study holds practical implication and can be used as a baseline study in the area of detecting automatic hate speech messages. Moreover, the output of different comparisons will be used as state-of-art techniques to compare future researches for existing automated text classification techniques.

A Seed-Guided Latent Dirichlet Allocation Approach to Predict the Personality of Online Users Using the PEN Model

“…Based on this factor, the SMO seems to be preferable because the classifiers registered insignificantly better predictions with reasonable time processing compared to the other three classifiers. The steady prediction of SMO agrees with the findings of previous personality recognition studies that have examined the representation of English messages [82,83]. As SMO aims to forward the Lagrange multipliers or alphas that satisfy the actual inherent learning process by identifying the support vectors [66], the transformation of inputs by the kernel function and optimization of subproblems minimized the computational cost for a large volume of the matrix.…”

Novel Hate Speech Detection Using Word Cloud Visualization and Ensemble Learning Coupled with Count Vectorizer

“…This kind of aggression-when it happens-in which a social media platform such as Twitter becomes a complex phenomenon, and many collaborative research fields are trying to solve this critical problem [6]. In the literature, many other works also have tried to address these issues, such as cyberbullying [6,7], trolling [8], extremism [9], hate speech [10,11] and racism [12,13]. There are different approaches that could be applicable for feature engineering, including N-gram to generate the vectors from dictionary of hate-related words [14][15][16].…”

An Interpretable Approach to Hateful Meme Detection

“…There has been a myriad of work on text-based hate speech detection, focused on Twitter-style text data. Current state-of-the art approaches [3,8,10] have involved the standard natural language processing toolkit, including BERT and other embedding schemes.…”