Communications of the ACM
A Better Way to Predict Election Outcomes
Boston University researchers have developed a way to correlate Web browsing patterns with public opinion polls, in order to generate more accurate predictions of election results.
Credit: iStock
Researchers at Boston University have developed a way to correlate Web browsing patterns with public opinion polls to create better predictions of election results.
The researchers obtained from media measurement firm comScore the Web browsing histories of more than 100,000 U.S. residents over the 56 days leading up to the 2016 election, and were able to determine when and where voters made decisions that led to the election of Donald Trump.
The team found support for Hillary Clinton began to decline on Oct. 25, 2016, three days before then-U.S. Federal Bureau of Investigation (FBI) director James Comey notified Congress that the bureau had found a new batch of relevant emails on the Clinton' server.
The researchers plan to further develop the system to work on encrypted data.
From Boston University
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found