The researchers' algorithm tries to detect the theme or topic of each story, and match it to other similar stories.

CNN: When bullying goes high-tech

Some of the new signals used by its algorithm are meant to detect bigness: the number of different bylines it publishes, the number of foreign bureaus it boasts, the sheer volume of copy it cranks out in a given day.

FORBES: Why Publishers Need to Stop Worrying and Learn to Love Google