Researchers from the University of Michigan in the US have developed a web-based tool to monitor fake news on social media platforms such as Twitter and Facebook.
The tool uses a Platform Health Metric called the ‘Iffy Quotient’, which draws data from two external entities: NewsWhip and Media Bias/Fact Checker.
How does it work?
• The NewsWhip is a social media engagement tracking firm, which collects URLs from hundreds of thousands of sites each day and then gathers information on, which of these sites have engagements on Facebook and Twitter.
• The Iffy Quotient queries NewsWhip for the top 5000 most popular URLs on the two social media platforms.
• Then the tool checks to see if those domain names have been flagged by Media Bias/Fact Check, which is an independent site that classifies various sources based on their reliability and bias.
• The tool divides the URLs into three categories based on the media bias and fact checklists:
Iffy: If the site is on the Questionable Sources or Conspiracy lists
OK: If the site is on any other list, such as Left-Bias, Right-Bias or Satire
Unknown: If it is not on any list.
The first report issued by the researchers confirmed what was suspected about the 2016 US presidential election, the Iffy Quotient increased dramatically on both Facebook and Twitter during the elections. It also showed that the two social media platform companies have made progress since early 2017 on their promises to crack down on misinformation, but one has succeeded more than the other. While the Facebook Iffy Quotient has seen a steady decline since early 2017 and has now returned to its early 2016 levels, the Twitter Iffy Quotient has not shown much decline and is still nearly twice its level in early 2016. Though Facebook and Twitter Iffy coefficients were roughly comparable through most of 2018, Facebook's is now somewhat lower. The contrast between Facebook and Twitter is even more pronounced in an engagement-weighted version of the Iffy Quotient, which can be taken as a rough proxy for the fraction of total user attention. |
-1756215887593.jpg)
Comments
All Comments (0)
Join the conversation