Hate Speech Detection using Logistic Regression on Bag of Words Model

Hate speech is a conundrum that is spreading like wildfire. Social networks allow members to connect by a variety of common interests and share information. These channels create and also help communities engage in discussing specific topics related to said news. However, they are also used as a medium to spread hate and offensive news. News on social media spread like a forest fire. Not only social media, but other forms of online media and precedence may cause hate speech to spread. Therefore, trying to prevent it beforehand is much better than causing an outrage. The objective of this paper is to try to establish a reliable model to identify hate speech in sentences. The purpose of this project is to analyze the sentiments of speech used sentences using certain ML algorithms. The Bag of Words model and Document Frequency Inverse Term Frequency (TFIDF) are used to process the text in sentences. Then we use the Logistic Regression algorithm on our bag of words.