Facebook will expand its fact-checking technology designed to combat fake news to both video and photo content.
Specifically, it has built a machine learning model that uses various engagement signals, including feedback from people on Facebook, to identify potentially false content. The photos and videos are then sent to moderators for review or resolution.
Facebook has 27 partners worldwide working to fact-check videos and photos to combat misinformation. Most of Facebook’s partner organizations have extensive experience in evaluating video and photo content, with expertise in image metadata analysis, reverse image search, and visual verification.
Fact-checkers, as the verifiers are called, can decide whether a video or photo shared by users on the social network provides false information, using all the capabilities described previously, thus helping Facebook train its machine learning model to improve the detection of misleading and incorrect content.
Facebook filters all videos and photos that are shared to be able to detect misleading content and classifies it into three different categories: fabricated, out of context, and misleading information.
