Figure 4From: Beating the news using social media: the case study of American IdolFor the last three eliminations we plot the ranking of each contestant measured as the percentage of tweets in the time window 8.00 PM-3.00 AM EST of the last three Wednesdays. We plot the 99% confidence intervals. In red we mark the contestant that was eliminated the next day. It is clear that even considering the errors, the ranking done considering the volume of tweets related to a specific contestant, is sufficient to identify the least preferred.Back to article page