Sentiment-Analysis_ICUP

Natural language processing for Twitter sentiment analysis using the Naive Bayes classifier algorithm to classify tweets as hate speech or not. The goal is to implement a Naive Bayes Classifier model and to be able to output an accuracy as well as to accept user input, to classify it into hate speech or not. We have been able to achieve this and our model classifies tweets with an accuracy of 93.62%. We also provide various data visualisations like bar graphs of the most commonly used hashtags and WordClouds of the most commonly used words. We train on 23000 tweets and test on 8964 tweets.