Why did Netflix think you'd like that?
It was predicted that Netflix would
spend around 5 billion dollars on programming in 2016, more than any other
company besides ESPN. So what would all this money go towards? Well for the
most part Netflix is investing in its recommendation algorithms it uses to
recommend TV shows and movies to its customers. These algorithms are important because
it is what keeps users engaged in the service and if customers don’t get good recommendations
they are more likely to stop using Netflix altogether. Therefore, it is very
important that the programs they use for recommendations are near perfect.
The way
Netflix recommends shows is by looking at the metadata for all the content the
service has. Tags like directors, actors, and genres can be assigned to
titles which the algorithm will then use to compare with titles the user recently
watched. So for example if I just watched Kill Bill, Netflix is likely to recommend
other titles directed by Quentin Tarantino. However, this is a pretty basic
form of recommendation and as many of us know, Netflix can have some very
specific categories for its movies. Netflix also sorts its programs based on
more specific identifiers. Shows that were made around the same time and shows
that received similar ratings are some more ways Netflix’s algorithm can
compare two titles. They also use humans to tag movies and tv shows with the
specific identifiers like you see in the photo below.
Before Netflix can recommend something
to you, it must first know what your tastes are. The company uses another set
of algorithms to track what you’ve watched and what you’ve searched for, what
your friends have watched, and even how you scroll through the list of titles available.
All of this is to create a profile of each users tastes in order to give better
recommendations to the user. Netflix, as well as other streaming services all
use computer algorithms to first find out what you’re interested in, and then
find other shows it thinks you will enjoy. The service is constantly improving
its programs and data gathering methods so that it can even better estimate
each customers tastes and be able to find the perfect movie for any occasion.
--------------------------------------------------------------------------------------------------------------
Sources:
http://www.bustle.com/articles/135173-the-7-best-genres-revealed-by-that-mind-blowing-netflix-hack-because-narrowing-the-options-isnt-easy/
No comments:
Post a Comment