Machine learning Spec. recommender systems week2

In the case of User 'Eve ’ who has not rated a single movie, how can user ratings be calculated when there is no similarity between the User ‘Eve’ and all the other users?
How is it feasible to use mean normalization to predict Eve’s ratings?

Does your question come from a lecture, or from a programming assignment?
Please give a reference.

My question is related to the content in the lecture

Which lecture? Can you give the title and a time mark?

Please place this thread in the week 2 category. You have posted in week 1.