Utilizing Support Vector Machines to Detect Hate Speech on Social Media

Abstract

This paper presents a novel application of Support Vector Machines (SVM) in developing automated systems to detect hate speech on social media platforms, addressing a critical need for scalable solutions to enhance online safety and societal cohesion. By leveraging SVM's proven efficacy in managing high-dimensional data and optimizing the balance between precision and recall, the study offers a comprehensive methodology that includes data collection, preprocessing, model training, deployment, and evaluation. The results demonstrate robust average performance across key metrics, affirming the model's reliability in accurately identifying hate speech while minimizing false positives. This research advances the field by showcasing the practical and theoretical contributions of SVM in automated hate speech detection, highlighting its potential to significantly improve content moderation practices. The findings underscore the necessity for ongoing refinement of detection systems and collaborative efforts among researchers, technology firms, and policymakers to create more inclusive online environments that promote respectful discourse and community well-being.

Keywords

• Automated Systems
• Classification
• Hate Speech Detection
• Sentiment Analysis
• Text Mining