Recommendation Systems

Sure. I do marathons…. on Netflix

In the digital world, recommendation systems play a significant role - both for the users and for the company/platform/sellers.

For the users, a new world of options are thrown up - that were hitherto tough to find. For companies, it helps drive up user engagement and satisfaction, directly impacting their bottom line.

If you’ve shopped on e-commerce platforms like Amazon or Flipkart, you would’ve seen options like:

“People who viewed this product also viewed…” “Products similar to this one…”

These are the results from recommendation systems. Netflix threw up a major data science challenge last decade: a million dollars to anyone who can improve their recommendation system by 10%. A recent estimate pegs the value of Netflix’s recommendation system to be worth $ 1 Billion.

In this full-day workshop, we will walk you through the various types of recommendation system. By the end of the workshop, you will have enough knowledge to build one for your problem.

Outline for the workshop

Session 1

What are recommendation systems?
- Definition
- Some examples
- Discussions around how to define the data for recommendation systems
Common types of recommendation systems
- Explicit Feedback
- Implicit Feedback
- item-based recommendation system
- user-based recommendation system
- Content-based recommendation system

Session 2

Collaborative Filtering (hands-on)
Matrix Factorization (hands-on)

Session 3

Content-based Recommendation System (hands-on)
Deep Learning in Recommendation System (theory only)

Session 4

Recommendation System in production
Deploying the recommendation system models
Measuring improvements from recommendation system
Improving the models based on the feedback from production
Closing thoughts and next steps

Pre-requisites

The workshop is approximately 50% theory and 50% hands-on.

Programming knowledge and basics of Python is necessary to follow the hands-on part.
No machine learning knowledge is assumed.
Basics of Linear Algebra will be good-to-have.

Software requirements

We will be doing this on cloud. Laptop with a browser is all you need for the workshop.

If you want to setup a local enviroment for it, the follow the next set of instructions.

We will be using Python data stack for the workshop. Please install Ananconda for Python 3 (https://www.continuum.io/downloads) BEFORE coming to the workshop.
After installing Anaconda, please install the following libraries for recommendation systems.

pip install lightfm
pip install scikit-surprise
pip install mlxtend

kaulmonish/recommendation