credibly & fake tweets on twitterNew research reveals what words sound phony on Twitter and what words increase credibility.  Researchers at the Georgia Institute of Technology scanned 66 million tweets linked to almost 1,400 real-world events to identify words and phrases that increase or weaken credibility on Twitter. Considering today’s anxiety over fake news, knowing and using words that increase credibility can add power to corporate communications.

“There have been many studies about social media credibility in recent years, but very little is known about what types of words or phrases create credibility perceptions during rapidly unfolding events,” Tanushree Mitra, the Georgia Tech Ph.D. candidate who led the research, stated in an article the school’s online news center.

Rating Credibility of Tweets

Researchers examined tweets surrounding events in 2014 and 2015, including the emergence of Ebola in West Africa, the Charlie Hebdo attack in Paris and the death of Eric Garner in New York City. They asked people to rate credibility of the tweets (from “certainly accurate” to “certainly inaccurate”). The team then fed the words into a computer program that split them into 15 different linguistic categories that included positive and negative emotions, hedges and boosters, and anxiety.

The program examined the words to judge if the tweets were credible or not. It matched the humans’ opinions about 68 percent of the time — significantly higher than the random baseline of 25 percent.

Tweets with “booster words,” such as “undeniable,” and positive emotion terms, such as “eager” and terrific,” were viewed as highly credible, Mitra said. “Words indicating positive sentiment but mocking the impracticality of the event, such as ‘ha,’ ‘grins’ or ‘joking,’ were seen as less credible. So were hedge words, including ‘certain level’ and ‘suspects’” she revealed.

Some Surprising Correlations

Surprising finding: Higher numbers of retweets also correlated with lower credibility scores. In addition, replies and retweets with longer message lengths were perceived as more credible.

“It could be that longer message lengths provide more information or reasoning, so they’re viewed as more trustworthy,” she said. “On the other hand, a higher number of retweets, which was scored lower on credibility, might represent an attempt to elicit collective reasoning during times of crisis or uncertainty.”

The system could eventually help combat fake news, an insidious problem on social media. The computer program could at some point evolve into an app that rates the perceived trustworthiness of an event as it unfolds on social media.

“When combined with other signals, such as event topics or structural information, our linguistic result could be an important building block of an automated system,” said Eric Gilbert, Mitra’s advisor and an assistant professor in Georgia Tech’s School of Interactive Computing. “Twitter is part of the problem with spreading untruthful news online. But it can also be part of the solution.”

Other Ways to Find Genuine Posts

Automated programs that attempt to root out fake news and kinds of false statements remain a work in progress. Facebook, widely accused of enabling the spread of fake news, recently announced it upgraded its algorithm to move to the top of users’ news feeds those posts that people consider genuine and not misleading, sensational or spammy.

Facebook will consider if the Page making the post regularly posts spam or if it tries to game feed placement by asking for “likes” or comments. It will also consider if its posts are often being hidden by people reading them and other factors.

Bottom Line: New research analyzes what prompts tweets to be perceived as credible or untrustworthy. Some of the findings may be surprising. While automated ratings of the credibility remains a work in progress, PR, marketers and event promoters may wish to consider the findings when promoting their messages on Twitter and other networks.