The track addresses context-aware recommendation and context-aware evaluation and live evaluation of recommendations. For this, a dataset from moviepilot is made available, this dataset contains information related to concepts from the world of cinema, e.g. single movies, movie universes (such as the world of Harry Potter movies), upcoming details (trailers, teasers, news, etc). At the end of the challenge, a live evaluation session will take place where algorithms trained on offline data will be evaluated online, on real users.
The aim is to find the right audience for a given movie. This movie does not necessarily exist already (as it might be in production), the goal is to generate a large impact on the recommended item in terms of interaction in the social networks of those users to whom the movies are recommended.
Update: The dataset is no longer available