MachineX: The Power of Recommendation Engines

Reading Time: 4 minutes

In this blog, we are going to talk about, what actually Recommendation Engines is and different types of same.

You can see the full webinar, related to this blog here :

Recommender Engines or Systems is one of the most mainstream utilization of data science today. They are utilized to predict the “rating” or “preference” that a user would provide for a thing. Pretty much every significant-tech organization has applied them in some structure or the other: Amazon utilizes it to suggest items to users, YouTube utilizes it to choose which video to play next on autoplay, and Facebook utilizes it to prescribe pages to like and individuals to follow. Furthermore, for certain organizations – think Netflix and Spotify, the plan of action and its success spins around the strength of their recommendations.

Netflix’s long list of suggested movies and TV shows is a fantastic example of personalized user experience. In fact, about 70 percent of everything users watch is a personalized recommendation, according to the company.

Getting to that point hasn’t been easy, and improving on its recommendation system is an ongoing process. Netflix has spent well over a decade developing and refining its recommendations.

In a very general way:

Recommendation Systems are algorithms planned for recommend revelant things to users (things being films to watch, items to purchase)

for example, I use to watch youtube when I am taking a break from my work. I decided not to watch Youtube for more than 5 to 10 minutes to rest my mind. Here is the thing that normally occurs, after I complete watching one video, the following video jumps out from Youtube recommendation and I click on that video. At the point when I take a look again at my watch, for reasons unknown, I have been viewing Youtube for over 60 minutes! :’)

There are 3 types of recommendation Systems

Simple Recommendation Engines

Suggesting all movies, which are in trending/new to all users. 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.

Content-based Recommendation Engines

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.

For example in the above screenshot, A user likes to read that red-colored articles. So, based on its content our engine will try to find out similar articles, maybe written by the same writer, or by the same publication or have a similar genre.

Collaborative-filtering Recommendation Engines:

Collaborative filtering is a technique that can filter out items that a user might like on the basis of reactions by similar users.

It works by searching a large group of people and finding a smaller set of users with tastes similar to a particular user. It looks at the items they like and combines them to create a ranked list of suggestions.

Collaborative filters do not require item metadata like its content-based counterparts.

There are two types of collaborative-filtering :

User-User based Collaborative-Engines

The main idea behind UB-CF is that people with similar characteristics share similar tastes.

The method identifies users that are similar to the queried user and estimate the desired rating to be the weighted average of the ratings of these similar users.

for example, In the above images, we can see that A cyclist preferred to eat a Pizza and salad. So, our Algorithm finds another person who also likes both of the items and in addition a cold drink. So we can recommend that cold drink or any item to the cyclist too.

Item-Item collaborative filtering

This method is quite stable in itself as compared to User-based collaborative filtering because the average item has a lot more ratings than the average user. So an individual rating doesn’t impact as much.

ITEM-ITEM collaborative filtering looks for items that are similar to the articles that the user has already rated and recommended most similar articles. But what does that mean when we say item-item similarity? It means how people treat two items the same in terms of like and dislike.

for example, In the above situation, User 1 likes to buy (orange, Apple, Banana ), whereas User 2 likes to buy (Orange, Apple). And then comes the user 3 who buy only (Apple), then our engine will see what other users buy with Apple. As a result, it will recommend Orange to User 3.

Conclusion

So, these were the types of recommendation engines. In the next blogs, We will make recommendation engines with content-based and collaborative-based techniques

Stay Tunes, happy learning 🙂

Resources

+ posts

Written by 

Shubham Goyal is a Data Scientist at Knoldus Inc. With this, he is an artificial intelligence researcher, interested in doing research on different domain problems and a regular contributor to society through blogs and webinars in machine learning and artificial intelligence. He had also written a few research papers on machine learning. Moreover, a conference speaker and an official author at Towards Data Science.

1 thought on “MachineX: The Power of Recommendation Engines5 min read

Comments are closed.