How Machine Learning is Overcoming Content Overwhelm in Streaming Video

Streaming video services are ubiquitous. Netflix, Amazon Prime, Instagram, and Facebook all boast their own versions. Last year, two billion users live-streamed across Facebook, creating huge amounts of content. Similarly, Netflix has 5,600 films and shows to watch on their platform. For consumers, the choice on these platforms is overwhelming.

It can be difficult to use the service to its full potential. How can a user find the content they want when so much is available? These platforms have anticipated this problem — and now machine learning is being used to assist users in finding satisfaction.

Firstly, machine learning can be used to interpret and condense long-form content to make it more useful and entertaining for viewers.

In the case of social media, machine learning detects how to react when a positive, negative or other result happens and has the ability to zoom, cut or overlay text depending on the trigger. Waffle Sport, a live-streaming video platform, streams sport on Facebook Live and uses algorithms to edit the coverage, for example to display replays of key moments during the game.

Secondly, machine learning can allow users to navigate these huge libraries and to make better choices.

Netflix uses machine learning to set recommendations for its users, related to watching history and flagged interests. Therefore, rather than overwhelming the user with thousands of films, Netflix can tailor their experience. Users are never going to be able to watch all of the shows on the platform, but will keep returning if they are able to have a manageable shortlist to work their way through.

Why is this interesting to Kitchen8? It’s a nice example of how new leading-edge technologies can be combined with each other to create powerful outcomes that are stronger than the sum of their parts. Streaming opens up huge possibilities for viewers, but introduces challenges — largely of decision paralysis. Machine-learning is softening those challenges and counterintuitively uses an algorithm to make the service feel more ‘humanised’