Content-Based Filtering
Recommends items similar to ones the user has already rated highly, with similarity measured on item features such as genre, language and length. It needs no data about other users.
When to use
Follow the four-step pipeline: describe each item with a feature vector, build a User Profile by averaging the vectors of liked items, compute Cosine Similarity to every unseen item, then rank. It handles new items well but lacks serendipity, since it only gives more of what the user already liked.
See Lec 6.