acm-header
Sign In

Communications of the ACM

ACM TechNews

Donald or Hillary? Tweetcast Predicts Your Vote


View as: Print Mobile App Share:
U.S. presidential candidates Donald J. Trump (left) and Hillary Rodham Clinton.

A new online tool can predict whether U.S. citizens will vote for Donald Trump or Hillary Clinton, based on their tweets.

Credit: Getty Images

Northwestern University researchers have developed TweetCast, an online tool that can predict whether citizens will vote for Donald Trump or Hillary Clinton based on their tweets with 80% accuracy.

Tweeting the words "lying," "liberal," "illegal," and "money," for example, indicates a user will vote for Trump. Using the words "single," "humanity," "rights," and "y'all," predicts a vote for Clinton. The researchers emphasize these are not the most prevalent terms voters use on Twitter, but rather they are the most predictive terms.

TweetCast uses a machine-learning algorithm to examine words, hashtags, tagged usernames, and mentioned websites to determine which terms are most predictive of voting preference.

Although the Northwestern researchers did not develop the algorithm used in TweetCast, they are the first to apply its approach to determining political preferences by analyzing tweets.

The algorithm was trained on Twitter users who have publicly declared support for one of the two candidates. During training, the algorithm identified patterns in those users' activity and applied the patterns to users across Twitter. The researchers also expanded the tool to predict which states each candidate will win.

"We can determine a lot from the language you use, including which restaurants you like, books you read, sports you enjoy, news you consume--and who you'll vote for," says Northwestern professor Larry Birnbaum.

From Northwestern Now
View Full Article

 

Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA


 

No entries found