Your task is to analyze an existing recommender system that you find interesting. You should:

  1. Perform a Scenario Design analysis as described below. Consider whether it makes sense for your selected recommender system to perform scenario design twice, once for the organization (e.g. Amazon.com) and once for the organization’s customers.
  2. Attempt to reverse engineer what you can about the site, from the site interface and any available information that you can find on the Internet or elsewhere.
  3. Include specific recommendations about how to improve the site’s recommendation capabilities going forward.
    Create your report using an R Markdown file, and create a discussion thread with a link to the GitHub repo where your Markdown file notebook resides. You are not expected to need to write code for this discussion assignment.

My Consideration here is:- Netflix recommender system

Scenario Design

What are the target users?
More than 65 Million users around the world who use netflix as their streaming service to watch movies and TV content. It provides personalized recommendation for each individual users.

What are their key goals?
Key goal is to provide personalized recommendation according to users likes and ratings. By doing this they better the user experience and more likely to keep the customers with them.

How can I help them accomplish their goals?
By providing correct recommendation vides to watch which are inline to customer insterests. So the user will watch more videos and spend more time on netflix.

Reverse engineer
The recommender system includes a number of algorithms. Some of the most important are:-
Personalized Video Ranker: PVR (Somewhat dependent on capabilities of the device)
Top-N Video Ranker (Finds best few personalized recommendations in the entire catalog for each member)
Trending Now
Continue Watching

There are many more of those algorithms but you can find all of it in the below link.

A full detail on netflix Recommender system is here:- https://dl.acm.org/citation.cfm?id=2843948
Additional Reference:- http://www.wired.co.uk/article/how-do-netflixs-algorithms-work-machine-learning-helps-to-predict-what-viewers-will-like

Specific recommendations for improvements

Some tags could still be wrong in netflix. Specially, if the movie is in foreign language i have observed that the recommendations sometimes lack that level of filtering capability in genres of foreign language movies.

I feel the article above has addressed this but would like to still mention Controlling for presentation bias. videos that members engage highly with are recommended to many members, leading to high engagement with those videos, and so on. Yet, most of our statistical models, as well as the standard mathematical techniques used to generate recommendations, do not take this feedback loop into account.