The New York Times announced it would open more articles to comments, and would start using a system, developed by Jigsaw, to identify comments that a human moderator would have probably rejected.
The tool, called Perspective, has been fed a corpus of Times comments that have been processed by a human moderator, teaching the algorithm what types of comments should be rejected, according to Nieman Lab. The system can be tried here.
According to Poynter, the new tool will allow the Times to increase the ratio of stories that are open to comments from 10 to 25 per cent. Ultimately, the newspaper hopes to push that number up to 80 per cent.
As internet companies are pressured to crack down on fake news and misinformation, it is likely that similar machine learning-based systems will be employed in the future to assess the credibility of content itself, not only comments.