Recommendation Systems

Mark Subra
2 min readAug 1, 2020

The proliferation of streaming sites, as well as YouTube, has led to data scientists coming up with recommendations for viewers. There is a reason why certain videos pop up in your YouTube or streaming platform’s homepage or app. The algorithm uses your past viewing history to come up with content you are likely to view.

By knowing your attributes such as age, location, etc., the recommendation system is able to deliver content that you are likely to click on. Advertisers are keenly aware of this, and they pay for ads between or before the video is played.

YouTube Ads

Recently, to the disdain of users, YouTube has been flooded with ads. They then offer a premium service without ads for a fee. This is all based on recommendation systems. The type of ads are often targeted based on the kind of video you are about to view. Data scientists are not perfect, and that is why you see unrelated ads to the type of video you are viewing. Why would Mr. Clean pop up during a cat video?

Imperfect as data science is, it still is valuable to the people who want to sell you products and streaming services.

Netflix

There are many streaming services, but the king is still Netflix. The platform has released several Emmy and Oscar winning programs. Netflix uses your information to recommend programs you are likely to view. There is a saying in finance, “Past performance is not an indicator of future performance.”

To a data scientist, this is the opposite. Netflix uses your history to figure out what shows it should put on your homepage as well as new shows to showcase. The work of data scientists goes to marketing and advertising and up the corporate ladder in order to get you to keep paying that monthly fee.

--

--

Mark Subra

I am a data scientist having recently graduated from the Flatiron School Immersive Data Science Bootcamp