- Recommender Systems (RSs) are software tools and tech-niques providing suggestions for items to be of use to a user. Often, better recommendations can be generated if the con-text of the recommendation is known, e.g., in a music RS, the user mood or activity. However, to adapt the recom-mendations to the context the dependency of the user pref-erences from the contextual conditions must be modeled. This requires explicit user evaluations/ratings for items in alternative contexts. In this work we investigate a novel approach for collecting and using contextually dependent ratings in recommender systems. We introduce the concept of "best context", i.e., the contextual conditions most suited for a particular item to be recommended. We designed an interface for collecting such data for music tracks. The col-lected data was then used to evaluate the quality of several "best context" prediction methods based on user-to-user col-laborative filtering. The results, in opposition to what we expected, show that the notion of best context is user depen-dent. Moreover, among the approaches we tried, the best performing one uses a k-nearest neighbors classifier where the user-to-user similarity measures the agreement of two users in assigning the best context to items.