From Words to Hate: Analyzing Predictive Features for Hate Speech Detection on X
Tushar Rawat, Saksham Bhatt, Daksh Rawat, Satvik Vats, Vikrant Sharma · 2024
Hate speech or hate comments are in the form of racism and sexism which is common on different social media platforms like X, Instagram, Facebook, etc. Due to these reasons, there are many industries and academics interested in hate speech detection on social media. There are many data sets which encourage the use of crowd sourcing for the annotation effort. In this paper, provided an examination of the brighter knowledge of hate speech on random forest model by obtaining from training on expert and amateur annotations. Provided an evaluation on our own data set and run on own model on the data set which various user’s social media ID’s and mentioning with their comments on various posts and this data set which contain different categories comments like hate speech, offensive languages, hate speech and offensive language, and non-hate and offensive language. And in this model, random forest algorithm is used which helps to classify the result or detect the user entered language because of these systems trained on expert annotations outperform system trained on amateur.