Detecting Hate Speech on Social Media Using Machine Learning

As times have progressed, the usage of social media has exponentially increased. Private and public opinions about a wide assortment of subjects are communicated and spread ceaselessly by means of various online social media platforms. Twitter is one of such platforms that has gained a lot of popularity. Twitter offers organizations and individual users a fast and effective way to advertise and communicate their ideas and thoughts without much hassle. Thus, analyzing customers' perspectives toward day to day events is crucial to success in the market place. Building up a program for notion examination is a way to deal with be utilized to computationally gauge people’s perceptions. This project applies sentiment analysis to a dataset containing thousands of tweets relating to a given string that is searched, all using R libraries. Searched strings could include hash tags, usernames, specific words etc. Using the processed output, we are able to determine the sentiments of people regarding any trending topic. Tweets are extracted using R and the data is wrangled by removing emoticons and URLs. Lexical Analysis is used to predict the meaning of tweets and subsequently infer the opinion graphically through ggplots, histogram, pie chart and tables.