Detection of Hate Speech on Social Media Using Machine Learning

Abstract

As we know, social media platforms are the fastest method of communication because information is transmitted and received virtually instantaneously. Social media platforms' development radically altered how people communicate in our world, and one effect of these changes is a rise in inappropriate behaviors, like the use of derogatory language online. It might be challenging to express harsh or disrespectful ideas to someone right in the face. People believe it is safe to abuse or post inappropriate content online. Hate speech can be detrimental to an individual or a community. The prevalence of hate speech on social media is widely acknowledged to be a serious issue. Techniques for detecting hate speech have received major investment from numerous governments and organizations, and they have also caught the interest of the scientific community. Since each proposed method has its own benefits and drawbacks, it is still challenging to evaluate how well it performs despite the abundance of literature on the subject. Filtering or blocking such content on the Web depends on its detection. Massive amounts of text data require manual processing and classification, which takes time and effort. To recognize this kind of content, however, automatic approaches are crucial because of the enormous quantity of data that is posted day by day. we used machine learning techniques like natural language processing, logistic regression, and random forest.