top of page
  • Writer's pictureDRL


Updated: Apr 6

Have you ever wondered how these personalized recommendations are developed across the internet? Have you ever wondered how Netflix suggests movies that you end up watching and liking? Have you ever wondered how Amazon shows you the perfect product that you end up buying? Well, the answer to all of these questions is Recommender Systems. These companies have a set of products and their attribute data or a set of product ratings from the users. This data is converted into a personalized recommendation system. E-commerce and retail companies are leveraging the power of data and boosting sales by implementing recommender systems on their web pages. The use of these systems has been gradually increasing for a few years and it’s a great time to dive deeper into this amazing machine learning technique.


The rapid growth of data mining techniques has led to a new era of information. More efficient techniques in data management systems have enabled better use of data. One such use of data is recommending more meaningful products and services to the customer. Recommender systems aim to provide a personalized recommendation to the user based on their preferences. In a way, recommender systems try to narrow down the choices of the user by presenting them with suggestions or choices that they are most likely to buy or use. Almost every major tech company has applied them in some form or the other: Amazon uses it to suggest products to customers, YouTube uses it to decide which video to play next on autoplay, and Facebook uses it to recommend pages to like and people to follow. Moreover, companies like Netflix and Spotify depend highly on the effectiveness of their recommendation engines for their business and success.


Companies using recommender systems focus on increasing sales as a result of very personalized offers and enhanced customer experience.

Recommendations typically speed up searches and make it easier for users to access the content they’re interested in and surprise them with offers they would have never searched for.

What is more, companies can gain and retain customers by sending out emails with links to new offers that meet the recipients’ interests or suggestions of films and TV shows that suit their profiles.

The user starts to feel known and understood and is more likely to buy additional products or consume more content. By knowing what a user wants, the company gains a competitive advantage and the threat of losing a customer to a competitor decrease.


On October 2, 2006, Netflix launched a competition called the Netflix Prize which was an open competition for the best collaborative filtering recommender system to predict user ratings for films, based on previous ratings without any other information about the users of films, i.e. without the users or the films being identified except by numbers assigned for the contest.

You can go to the following link for further information: THE NETFLIX PRIZE


Every company has its own need to use and implement Recommender Systems. While someone needs this to improve their sales others might use this to simply make the work more efficient. Few companies implement this using entire historical data as a basis whereas others like to implement more intelligent versions. Based on the requirements, recommendation engines can be broadly classified into 3 categories:

4 (A): Simple recommenders or Demographic Filtering:

Simple recommenders offer generalized recommendations to every user, based on movie popularity and/or genre. The basic idea behind this system is that movies that are more popular and critically acclaimed will have a higher probability of being liked by the average audience. An example could be IMDB Top 250. Similarly, products which generate more revenue or have more frequent purchase are recommended to the customers.

4 (B): Content-based recommenders

Content-based recommenders suggest similar items based on a particular item. This system uses item metadata, such as genre, director, description, actors, etc. for movies, to make these recommendations. The general idea behind these recommender systems is that if a person likes a particular item, he or she will also like an item that is similar to it. And to recommend that, it will make use of the user's past item metadata. The entertainment industry relies heavily on such systems, for example, YouTube and suggest new videos or songs based on the historical use of the application and the metadata associated with it.

4 (C): Collaborative filtering recommenders

Collaborative filtering recommenders these systems are widely used and they try to predict the rating or preference that a user would give an item-based on past ratings and preferences of other users. Collaborative filters do not require item metadata like its content-based counterparts. Collaborative filtering based recommenders can be further divided into two categories:

1. Memory Based Recommenders :

Memory-based algorithms approach the collaborative filtering problem by using the entire database. Memory-based recommender tries to find users that are similar to the active user (users we want to make predictions for) and uses their preferences to predict ratings for the active user.

2. Model-Based Recommenders :

Model-based Recommenders use a complex machine-learning algorithm to estimate the ratings. A typical example is the singular value decomposition of the user-item rating matrix or the K-Nearest Neighbours based model.



It is difficult to explain all the topics in one post. This blog is a part of the series on the Recommendation Systems where we will learn to implement them using Python or R. Read other posts for a more in-depth understanding of the mathematics behind this excellent machine learning technique.

41 views0 comments

Recent Posts

See All
bottom of page