Movie/Show Recommendation System

Lesson 35 of 7410 min readPremium lesson preview

Movie/Show Recommendation System

Use the questions below to review the whole recommendation pipeline. In the first answer, show how candidates from different sources reach a common ranker and then a final list. In the second, remove the viewing history or the item’s interaction history and explain what evidence remains. In the third, test whether your proposed objective can reward a result that the viewer would consider poor.

Before reading each reference answer, write one concrete example that could expose a weakness in your design. A popular but irrelevant title, a new movie with no watches, or a long session with low satisfaction should lead you to different changes.

The rest of this lesson is for Premium members

Premium opens every lesson in this course, along with all questions and solutions on PracHub.