An Algorithm to Limit Inaccurate Tweets

January 10, 2013

Since its famous role in the Arab Spring, Twitter‘s status as an active participant in (as opposed to simply a documenter of) unfolding events has been self-evident. Since then, on several notable occasions, users of the service have supplied crucial information before traditional news sources got their hands on the facts. However, as we saw during the tragic events of December 14, sometimes Twitter users get it wrong. Sometimes, the misinformation causes unnecessary stress, confusion, and even danger. That’s quite a downside to the otherwise helpful contrivance. What is a concerned citizen of the world to believe?

A solution may be on the way. It is after the fact (this time), but it is progress nevertheless. Slate’s “Building a Better Truth Machine” examines the possibility that machine-learning algorithms could identify and halt false rumors before they pervade the Twittersphere. Several studies have recently emerged that identify common characteristics of both true and false tweets. (See here and here for a couple of examples supplied by the article.) Writer Will Oremus tells us:

The authors of the 2010 study [from Yahoo Research, here] developed a machine-learning classifier that uses 16 features to assess the credibility of newsworthy tweets. Among the features that make information more credible:

  • Tweets about it tend to be longer and include URLs.
  • People tweeting it have higher follower counts.
  • Tweets about it are negative rather than positive in tone.
  • Tweets about it do not include question marks, exclamation marks, or first- or third-person pronouns.

Several of those findings were echoed in another recent study from researchers at India’s Institute of Information Technology who also found that credible tweets are less likely to contain swear words and significantly more likely to contain frowny emoticons than smiley faces.

Interesting. Oremus admits that those looking to purposely spread lies are sure to find a way around any algorithm that may be put in place, but suspects that it would at least cut down on the proliferation of inaccuracies. Let us hope that he is correct, and that an effective solution is implemented soon.

Cynthia Murrell, January 10, 2013

Sponsored by, developer of Augmentext


Comments are closed.

  • Archives

  • Recent Posts

  • Meta