NCSU program samples Twitter sentiment during debates
Posted October 17, 2012
Raleigh, N.C. — WRAL partnered with Associate Professor Christopher Healey of the North Carolina State University Department of Computer Science to develop and deploy a tool to track what Twitter users had to say during the dual debates Tuesday night.
Healey's program shows how many Tweets use the name of the candidate or candidates and the overall sentiment – positive or negative – contained in the words.
The program uses a "sentiment dictionary" to correlate words used in tweets with emotions. The program estimates how intense someone is feeling as well as whether their language is more or less pleasant.
After the debates ended, Healey was able to capture the window of time during which people watched. Users can zoom in on a moment during the debate and even search Tweets by keyword.
A word of warning: There were millions and millions of Tweets during the presidential debate, so the program can take some time to load even on a fast connection.
"Twitter gives you a really nice way to capture people's emotions at the point and time the debate is unfolding," Healey said. Unlike blogs or even Facebook, Twitter users tend to post about what is happening in front of them, providing a sample of real-time opinions. The effect is not dissimilar from a focus group "dial test," in which a small group of voters is monitored for their reaction to candidates.
Some observations from Tuesday night's debates:
- The volume of Twitter traffic regarding the gubernatorial debate was low, reflecting a low level of engagement.
- During the presidential debate, there were more tweets about Romney than Obama.
- Certain points in the debate caused spikes in traffic, including discussions about pay-equity for women, the terrorist attack in Libya, the exchange between Mitt Romney and Barack Obama over pensions, and Obama's closing remarks that referenced the 47 percent.