Scott Clement of The Washington Post writes that researchers at the Laboratory for Social Machines have found that while the majority of Twitter conversation concerning the presidential campaign has centered around Donald Trump over the past week and a half, “battlegrounds differed in what particular issues or themes they focused on.”
MIT researchers have found that immigration has been dominating election conversations on Twitter, writes John West in a Washington Post article. “Tweets on immigration soared to almost 60 percent of the election-related Twitter conversation after Donald Trump’s statements about a potential 'softening', his visit to Mexico and then his address on the topic.”
Using data compiled by the Media Lab’s Laboratory for Social Machines concerning Twitter conversation before and after the mass shooting in Orlando, Aaron Blake of The Washington Post shows that political priorities leading up to the presidential election “may depend heavily on world and domestic events that nobody can predict.”
In an article for Reuters, James Saft writes that MIT researchers have found that analyzing Twitter sentiment can provide useful information for investors. “We exploit a new dataset of tweets referencing the Federal Reserve and show that the content of tweets can be used to predict future returns,” Prof. Andrew Lo and grad student Pablo Azar explain.
MIT researchers are creating a GIF genome that will allow computers to identify the emotions or meaning behind a GIF, writes Kevin Maney for Newsweek. “We realized that they’re becoming more and more serious of a medium,” explains graduate student Kevin Hu. “And we realized that we could quantify this usage.”
Terri Rupar reports for The Washington Post that researchers from MIT’s Laboratory for Social Machines have analyzed Twitter conversation surrounding the Supreme Court vacany and found that “people are definitely seeing the vacancy and Obama's nomination as issues for the 2016 election.”
In an article for CNN about Instagram accounts that highlight scientific developments, Esra Gurkan features the MIT account. Gurkan writes that the MIT Instagram account combines “both science and beauty, providing unique views of the amazing architecture found on their campus and, of course, the quirks and ingenuity of being a student there.”
In this BBC News video, postdoc Brad Hayes explains how his algorithm uses transcripts of presidential candidate Donald Trump’s speeches to compose tweets. “The real reason why this works is because Donald Trump tends to use these very short, imperative statements,” Hayes says.
CBS News reporter Brian Mastroianni writes about how social media users celebrated Pi Day this year, highlighting two of MIT’s Facebook posts. One of the MIT Pi Day posts featured MIT students writing down as many digits of π as possible from memory, while the other wished Albert Einstein a happy birthday.
Washington Post reporter Scott Clement compares the results of an analysis performed by MIT researchers of key issues on Twitter in the 2016 presidential race to a national survey. The researchers found that foreign policy and race are key issues on Twitter, while the national survey found that the economy and jobs were top priorities for voters.
New York Times reporter Penelope Green speaks with Prof. Sherry Turkle about a new Facebook tool aimed at making breakups easier. “It’s not to say that Facebook shouldn’t make it easy to click that button to avoid certain painful memories,” she said. “But the reason we’re looking through those old love letters is we’re trying to work through our past.”
Matt McMarland writes for The Washington Post that a CSAIL researcher has developed a computer system that can produce tweets that read like they are written by presidential candidate Donald Trump. McFarland explains that postdoc Brad Hayes “wanted a fun way to familiarize himself with some statistical modeling techniques for his research on human and robot interactions.”
Popular Science reporter Dave Gershgorn writes that postdoc Bradley Hayes has created an algorithm that tweets fragments of Donald Trump’s speeches. Gershgorn explains that, “the algorithm works by selecting a random letter, and then predicting what letter would normally come next, based on the original text.”
MIT researchers have found that Donald Trump and Hillary Clinton are the leading online influencers of the 2016 presidential election, reports Natalie Andrew for The Wall Street Journal. Research scientist William Powers explained that the findings show, “how influential social media and earned media is in the election.”
Researchers from the Laboratory for Social Machines have partnered with The Washington Post to examine how people are discussing the current presidential election on Twitter. The researchers use a program that detects and categorizes tweets “to see which issues or candidates have had the biggest share of the conversation.”