The previous sections analyzed the content of tweets. In this section, we will discuss another interesting aspect of analyzing data from Twitter-the distribution of tweets over time.

Generally speaking, a time series is a sequence of data points that consists of successive observations over a given interval of time. As Twitter provides a created_at field with the precise timestamp of the tweet, we can rearrange tweets into temporal buckets so that we can examine how users react to real-time events. We are interested in observing how a population of users is tweeting, not just a single user, so the data gathered via the Streaming API is most suited for this type of analysis.

The analysis in this section uses the dataset from the 2015 Rugby World Cup Final. This is a nice example of how users react to real-time events such as sport events, concerts, political elections, and everything from major disasters to TV shows. Other applications of time series...