Picking What to Watch Next: Build a Recommendation System
summary
Builds an item‑based collaborative filtering recommender by turning the dataset into an m×n rating matrix, filtering low‑activity users and unpopular items, converting to a CSR sparse matrix, and fitting a k n nearest neighbor model using cosine similarity. Uses pandas for prep, a MovieLens CSV download as the movie lens data set, fuzzy title matching, and a cached pickle for fast browser demos via pi script. Notes popularity bias and cold‑start; code has an unfinished exercise branch and a finished branch.
AI-written summary · about
chapters (10)
- 0:00Workshop introduction
- 2:37Recommender system basics
- 6:37Methods of filtering
- 9:31Data acquisition and setup
- 12:38Data preparation and matrix
- 15:55Model building and training
- 18:53Code structure and setup
- 21:55Implementing the recommender
- 26:51Model inference and testing
- 29:19Conclusion and resources
related talks
37mA primer on recommendation systems.
01h 31mHands on - Build a Recommender system
01h 38mBuilding Recommender Systems Using Python
32mLet's talk about Recommendation Systems!
20mBuilding Recommender system
46mBeyond Bag of Words A Practitioner’s Guide to Advanced NLP
43mBuilding your first recommender
15mWhat they didn’t teach you in Grad School #bbuzz