Hate Speech Detection on Twitter: A Machine Learning Approach to Identify Attacks on Indigenous People During the 2022 Ecuador Strike

The exponential growth of social media platforms has facilitated worldwide message dissemination and diverse information exchange in real-time. However, this communication diversity can also foster hate speech due to varied opinions. This research delves into detecting hate speech during the June 2022 national strike in Ecuador, where derogatory comments targeted indigenous protesters on Twitter. The study aims to comprehend the prevalence and nature of hate speech by implementing a system to detect tweets into “hate speech” and “non-hate speech” classes using a Multinomial Bayesian Model. The application’s performance is evaluated using different metrics related to the prediction accuracy, obtaining a value of 0.84. The results provide valuable insights into the dynamics of hate speech during significant events, underscoring the significance of analyzing tweets to counter hate speech on social media platforms and encourage a more inclusive and respectful online conversation.