• This model scraps Twitter so as to scan for tweets.
• After web scraping (with the help of Twint and Tweepy modules in Python and Beautiful Soap and Selenium and Twitter API) we make our .csv file.
• After the .csv file is prepared, we need to preprocess the data and clean the data and distinguish and convert the continuous data into categorical data.
• Then we will divide our dataset in testing and training set for training and testing of our model.
• Then we need to apply Machine Learning Models (such as Support Vector Machines, Logistic Regression, Vectorization, Decision Tree Classifiers, Neural Networks) to detect the presence of hate tweets.
• After the detection, we need to analyze the result.
• We will make a Google extension so that it works on every website in Google. Our extension directly connects the Twitter website to our Machine Learning model and gives the output.
• If a hate tweet is found, our model flashes a warning to the user.
• If the Twitter website has no such hate tweets, it launches smoothly