Building Scalable Recommendation Systems

Nov 16, 2019

09:30

Prerequisites

The workshop is having both theory and hands-on parts. 1. No machine learning knowledge is required 2. Basics of Python Programming are necessary to follow the hands-on part.

Recommendation Systems are the backbone of some of the most awesome products we see around today. If you have done shopping on Amazon or Flipkart you may have come across things like:

  • People who bought this product also bought
  • Products recommended for you

Whether watching videos of Youtube, buying food online, listening to your favorite songs on or finding your loved one on Tinder, you certainly would have hit some of their recommendation systems. For users these systems help in finding things they may like and for companies it increases user engagement and satisfaction which directly impacts their bottom line

Apart from these, recommendation systems are used in a plethora of industries like fintech, retail, E-commerce and even search

Have you ever thought how do they generate these precise recommendations in real time for millions of users?

In this workshop we will answer this question where we will take case studies from different industries, refine the business problem to a Machine Learning problem and show how it is scaled up to work for millions of users. We will try to do everything from scratch so you get a deep understanding of these systems

Topics we will be covering in the workshop:

  • What are Recommendation Systems?
    • Definition
    • Real life examples
    • Defining data for recommendation systems
  • Fundamentals of Machine Learning and Linear Algebra
    • Types of machine learning algorithms
    • The row and column picture
    • Dimensionality Reduction
  • Collaborative Filtering Approaches
    • User – item approaches
    • Item – item approaches
    • Dimensionality Reduction in Collaborative Filtering
  • The Cold Start Problem
  • Content Based Approaches
  • Hybrid approaches
    • Mixing Collaborative Filtering and Content based approaches
  • Evaluation Metrics
    • Theoretical: Accuracy, Precision, Recall, ROC
    • Practical Evaluation Metrics – Business Metrics, A/B Testing, Google Analytics
  • Association Rule Mining and Market Basket Analysis
  • Building APIs and scaling them up
  • Search as a Recommendation system
  • Deployment of these systems

Case Studies we will be covering in the workshop:

  • Building a job recommendation system for a jobs portal
  • Video recommender system similar to YouTube
  • Optimizing product placement in Retail to increase sales
  • Food discovery similar to Uber Eats

Hands on projects we will be doing in the workshop:

  • Building a movie rating recommendation system (similar to Netflix)
  • Building a Song Recommendation System
  • Building a Semantic Search system

BUY NOW

  • Anand Mishra

    Head of Engineering

    Analytics Vidhya

    BIO

    Anand Mishra is Head of Engineering at Analytics Vidhya. He is an entrepreneur, an engineer and a data science professional all rolled into one. He co-founded MudraCircle, the true lending marketplace leveraging machine learning to fulfill SME loans. Before MudraCircle, Anand has worked across several companies like Lendingkart, HTMedia as Head of Data Science, Tickled

Copyright 2019 Analytics Vidhya. All rights reserved