Paper mill cancer studies get double the number of citations as genuine papers

Part of the problem is how to weight citations. A citation in a good journal is better than a citation in a bad journal that is better that a link in X/Tweeter.

Someone tried to use PageRank to weight citations/impact index a few years ago. The problem was that Nature and Science were not the first two in their rank so they have to cheat and add some other weight to make the index look as expected.