Content-Based Filtering
A recommendation approach that suggests items similar to what a user has previously liked, based on item features rather than other users' behavior.
How It Works
Extract features from items (genre, keywords, embeddings). Build a user profile from features of items they've interacted with. Recommend new items whose features match the user profile.
Pros and Cons
No cold-start problem for new items (features are known immediately). Can explain recommendations ('because you liked sci-fi'). But creates filter bubbles and can't discover unexpected interests like collaborative filtering can.