Facebook has announced that it is expanding its fact-checking for photos and videos to all of our 27 partners in 17 countries around the world. According to the social media giant, this will help identify and take action against more types of misinformation, faster. The company revealed that it has built a machine learning model that uses various engagement signals, including feedback from people on Facebook, to identify potentially false content.
How it works is that, Facebook will send photos and videos to fact-checkers for their review who have expertise in evaluating photos and videos and are trained in visual verification techniques, such as reverse image searching and analyzing image metadata, like when and where the photo or video was taken.
The company said, it is using other technologies to better recognize false or misleading content. It said it is using an optical character recognition (OCR) to extract text from photos and compare that text to headlines from fact-checkers’ articles.
We are also working on new ways to detect if a photo or video has been manipulated. These technologies will help us identify more potentially deceptive photos and videos to send to fact-checkers for manual review.
Fact-checkers can assess the truth or falsity of a photo or video by combining these skills and using other methods such as research from experts, academics or government agencies. The tech giant revealed that misinformation in photos and videos usually falls into three categories which include,
- Manipulated or Fabricated,
- Out of Context, and
- Text or Audio Claim.
Facebook said that as it gets more ratings from these fact-checkers, it will get the chance to improve the accuracy of its machine learning model.
Source: Facebook News Room