3 ms·
I would imagine some kind of a/b testing algorithms have come up with "wildly misleading headline" => more interest.
by m463 6y ago
I would imagine some kind of a/b testing algorithms have come up with "wildly misleading headline" => more interest.
- eru 6y agoIf you assume that interestingness of headlines is somewhat randomly distributed, and that for each article there are more misleading headlines than accurate headlines, your relation is bound to be true, even if being misleading by itself doesn't help.