I have implemented solution for sentiment analysis on COVID-19 tweets.The training dataset consists of 41157 tweets and the testing dataset consists of 3798 tweets. In the original datasets there are 5 sentiments in the sentiment column namely positive, extremely positive, negative, extremely negative and neutral. In this project I have used 3 class classification for faster and easier analysis. The use of different models Logistic Regression , DecisionTree classifier , Naive bayes classifier , RandomForest Classifier is also demonstrated.