Sentiment analysis during Hurricane Sandy in emergency response

Sentiment analysis has been widely researched in the domain of online review sites with the aim of generating summarized opinions of users about different aspects of products. However, there has been little work focusing on identifying the polarity of sentiments expressed by users during disaster events. Identifying such sentiments from online social networking sites can help emergency responders understand the dynamics of the network, e.g., the main users' concerns, panics, and the emotional impacts of interactions among members. In this paper, we perform a sentiment analysis of tweets posted on Twitter during the disastrous Hurricane Sandy and visualize online users' sentiments on a geographical map centered around the hurricane. We show how users' sentiments change according not only to their locations, but also based on the distance from the disaster. In addition, we study how the divergence of sentiments in a tweet posted during the hurricane affects the tweet retweetability. We find that extracting sentiments during a disaster may help emergency responders develop stronger situational awareness of the disaster zone itself.