The Facebook's new algorithm that can guess what posts will be "most informative".
Surpurisingly, this algo is based on old-school way, survey.
This algorithm is using huge data feature, about 100,000 features.
Facebook did not tell us how they make these survey and how it affects things such as political bias, for instance.
Recently, Facebook makes effort to this kind of algorithms.
Facebook Inc is changing its news feed algorithm that can guess what posts will be "most informative" based on polling of people around the world.
a sampling or collection of opinions on a subject, taken from either a selected or a random group of persons, as for the purpose of analysis.
As usual, Facebook doesn't explain their algo that make ranks of feeding, but it did. It is the old-school approach, survey.
the rank one is really not informative and rank five is really informative
Generally, people find stories informative if they are related to their interests, if they engage people in vroader discussions, and if news related to the world around them.
Facebook did'nt immediately reply to our inquiry.
The feedback from these surveys is combined with how users interact with on Facebook. The factors that can be used in this algorithm is about 100,000 different features.
What is the informative is "different for each person and will likely change over time"
Then, most of users can not notice the difference of news feed by this algorithm, but this is first thing any Facebook user sees, it is not in any namesake app or website.
the act or process of altering
someone or something that has the same name as another person
the nost pages won't see any significant changes
to realize beforehand; foretaste or foresee
an act of referring
These changes follow other tweaks Facebook announced las week, all intended to de-emphasize what it considers "click-bait" post that exaggerate or mislead users.
tweaka crystal methamphetamine-like substance