A comparison of classification algorithms for hate speech detection

Abstract

Freedom of opinion through social media is frequently affect a negative impact that spreads hatred. This study aims to automatically detect Indonesian tweets that contain hate speech on Twitter social media. The data used amounted to 4,002 tweets related to politics, religion, ethnicity and race in Indonesia. The application model uses classification methods with machine learning algorithms such as Naïve Bayes, Multi Level Perceptron, AdaBoost Classifier, Decision Tree and Support Vector Machine. The study also compared the performance of the model using SMOTE to overcome imbalanced data. The results show that the Multinomial Naive Bayes algorithm produces the best model with the highest recall value of 93.2% which has an accuracy value of 71.2% for the classification of hate speech. Therefore, the Multinomial Naïve Bayes algorithm without SMOTE is recommended as the model to detect hate speech on social media.

The Demonization of Islam through Social Media: A Case Study of #Stopislam in Instagram

“…Social media contribute to freedom of speech, allowing people to express their thinking and share their feelings. However, some people abuse this and send offensive comments that could negatively affect the people [23]. The most damaging impact of freedom of speech is the number of messages full of hate shared by unconscious people.…”

“…The most damaging impact of freedom of speech is the number of messages full of hate shared by unconscious people. According to Putri et al, (2020) [23], hate speech "is any kind of communication which is offensive, underestimating, and humiliating an individual or group of people" (p. 1). As can be observed, the consequences of demonization and hate speech in social media cause: Racism, Islamophobia, and invisibility.…”

“…It should be noted that this research is a case study, so the research questions are analyzed through the #StopIslam hashtag. Besides, as a clarification of Research Question 3 (RQ3), it is important to emphasize that we are studying whether the promotion of the consequences of demonization mentioned by Putri et al, (2020) [23] is observed through the #StopIslam hashtag, not the effect on society of the analyzed posts.…”

“…Although Instagram does not detect it as illegal material because it does not promote actions against the law, it does observe discursive attacks on Muslims through the publication of messages that promote false ideas about Islam (e.g., Figure 4). This, according to some authors such as Putri et al, (2020) [23], can be considered as hate speech, so the Examples from the category A war between Muslims. Caption: People support criminals and stop supporting the French.…”

“…Although Instagram does not detect it as illegal material because it does not promote actions against the law, it does observe discursive attacks on Muslims through the publication of messages that promote false ideas about Islam (e.g., Figure 4). This, according to some authors such as Putri et al, (2020) [23], can be considered as hate speech, so the Instagram policy should be reviewed. Poole (2020) [52] also finds this very concerning, because it illustrates how social media platforms' strategies can create conditions that lend themselves not just to the actions but ideological commitments of right-wing populist groups.…”