Netflix: 67% rented movies are from Recommendations
Google News: 38% more click-through rates are due to recommendations
Amazon: 35% sales are from Recommendations
When was the last time you watched a movie at a theatre? How about the last time you watched a movie at Netflix, Prime Video or Hotstar? If you’re like most consumers, you’ve done the second thing much more often.
But how much power do platforms like Netflix, Amazon, or WeChat, have on consumers and your potential customers? If you have spent any time on these sites you will know they are full of ‘personal’ recommendations.
These online recommendations are based on algorithms which generate a series of suggestions for items that users might be interested in, and are based on previous choices.
For example, clicking and reading about ‘management’ generates recommendations for other articles about ‘management’. Using these algorithms in theory keeps consumers on the site by giving them more of what they want.
As a product manager optimizing for revenue, good personalised recommendations can go a long way in effectively getting that extra revenue with minimal effort. Now that we understand the effectiveness of these tools in the industry, let's look at the what will you learn in this course
Understand the importance of Recommender Systems in industry
Detailed Taxonomy of types of Recommender Systems
Collaborative Filtering Methods
Content Based Recommender Systems
Knowledge Based & Hybrid Recommender Systems
Market Basket Analysis & Association Rules
Evaluation of Recommender Systems
4 real life projects
- What is a recommender System?
- Industries where recommender systems are super relevant
- Structure of Course
- Instructor Introduction
- Introduction to Non Personalised Recommender Systems
- Why Non Personalised Recommender Systems?
- Weak Personalisation
- Case Study (Creating a Movie Recommender using summary stats)
- What is association Rule Mining?
- Market Basket Analysis & Brute Force Method
- Apriori Algorithm
- Solve Market Basket Analysis problem using Apriori Algorithm in Python
- Personalised Recommender Systems
- Types of Personalised Recommender Systems
- What is Collaborative Filtering?
- Implicit vs Explicit Ratings
- Steps for User Based Nearest Neighbour Collaborative Filtering
- Similarity Measures
- Generating Predictions
- Implementation on movielens data
- Memory Based vs Model Based Approaches
- Item Based Nearest Neighbour Collaborative Filtering
- Data Sparcity Issues
- Implementation for Item Based Collaborative Filtering
- Motivation & Intuition
- SVD Primer
- User & Item Feature Matrices
- Limitations of Collaborative Filtering Methods
- Evaluation of Recsys and difficulties
- How to select the right metric?
- Predictive Accuracy Metrics
- Classification Accuracy Metrics
- Rank Aware Metrics
- Metrics Beyond Accuracy
- Implementation of metrics in Python
- Introduction to CB Based Filtering
- Content Representation and Item similarities
- Term Frequencies & TFIDF
- User & Item Profiles for Content Based Filtering
- Implementation of CB Based Filtering
- Limitations of Content Based Filtering Methods
- Introduction to Knowledge Based Recommender Systems
- Case Based vs Constraint Based Recommender Systems
- Hybrid Recommender Systems
- Types of Hybrid Recommenders
Ankit is an IIT Bombay Graduate with a Masters and Bachelors in Electrical Engineering. He is a corporate trainer and leads the hackathon category at Analytics Vidhya. He is responsible for liaison with various companies to transform their data into data science competitions. He has conducted corporate trainings for a BFSI client on Basic and Advanced Machine Learning. He has finished in top 5 of multiple data science competitions and also conducted a workshop on how to win data science competitions at DataHack Summit 2019. He has previously worked as a lead decision scientist for Indian National Congress deploying statistical models (Segmentation, K-Nearest Neighbours) to help party leadership/Team make data-driven decisions. His motivation lies in putting data at the heart of business for data-driven decision making.