Its Friday night and my friend said she would like to watch a movie released today. I don't want to watch that movie so I have to convince her to watch football match instead. My friend said you have time till morning to convince me. Being a web developer, I decided to build a web application that would parse the tweet stream about the movie and give us the overall sentiment about the movie. If more that 60% people have positively tweeted about the movie then we would watch the movie otherwise we would watch the football match. To help others like me, I decided to host it in the public cloud so that I can share the application URL with others. In this session, you would learn how to build a web application that would parse the incoming stream of tweets using Python's NLTK. This session would start with introducing the problem space, then covering fundamentals and usage of sentiment analysis, and then looking at the NaiveBayesClassifier that does the job of classifying tweets into different categories. We would also talk about an open source Python library called TextBlob that builds on top of NLTK and provides rich API for text processing. The web application would be built using Python Flask web framework.