This document describes how to analyze tweets in real-time to visualize the emotional state of each US state. It involves using the PubNub data stream API to consume tweets with geo-location data and classify them as positive or negative based on keyword analysis. A dictionary of positive and negative words is used to analyze each tweet and tally the sentiments for each state. D3.js renders a TopoJSON representation of the US map and colors each state to reflect the prevailing sentiment based on the live tweet analysis and tally.