Movie recommender from viewer ratings
Inspired by publicly reported work at Netflix
Recommend films to each viewer from what similar viewers rated, and check the suggestions are any good.
The brief
Netflix has written in public about how much of what members watch comes from recommendations. Build a recommender from a large public ratings dataset: start with a popularity baseline, add collaborative filtering, and measure whether your model ranks the films a viewer liked higher than the baseline does. Good looks like: a fair train/test split by time, ranking metrics (precision@10, NDCG) against the baseline, and a short demo for three sample viewers.
Read the case first
What Netflix has said in public about this kind of work, on its own pages:
- Artwork Personalization at Netflix (Netflix Tech Blog) ↗
- Recommendations research area (Netflix Research) ↗
Make it your own
Pick an audience Netflix's posts don't cover: regional-language films, a college film club or short documentaries. Decide what "a good suggestion" means for them (new to them, in their language, not just popular) and measure that, not only accuracy.
AITeky has no link with Netflix, which has not reviewed or endorsed this brief. The brief only draws on what the company has published, uses open data rather than its data, and is a learning exercise: don't present your project as Netflix's work or system.
What you will build
- Explore ratings by viewer, film and time
- Build a most-popular baseline
- Train matrix factorisation or item-to-item similarity
- Evaluate with precision@10 and NDCG on a time-based split
- Show top-10 suggestions for sample viewers and explain them
Free, open data to build it with
Sign up free to get the data and links
It takes a minute with your email. AiTeky Pass (₹999/month or ₹9,999/year) is free for your first 30 days.