In the Mood4: recommendation by examples

Traditional recommender systems generate personalized recommendations based on a profile that they create for each user. We argue here that such profiles are often too coarse to capture the current user's state of mind and desire. For example, a serious user that usually prefers documentary features may, at the end of a long and tiring conference, be in the mood for a lighter entertaining movie, not captured by her usual profile. As communicating one's state of mind to a system in (key)words may be difficult, we present in this demo Mood4 - a novel plug-in for recommender systems, which allows users to describe their current desire/mood through examples. Mood4 utilizes the user's examples to refine the recommendations generated by a given recommender system, considering several, possibly competing, desired properties of the recommended items set (rating, diversity, coverage). The system uses a novel algorithm, based on a simple geometric representation of the items, which allows for efficient processing and the generation of suitable recommendations even in the absence of semantic information.