Watching a comedy can help a user escape from the negative mood, which in turn affect the user’s feedback over the movie. In other words, a non-cognitive mood inducer (the movie) can affect a user’s post-consumption evaluation over the inducer (the rating the user give) which is directly associated with users’ assessment over consumed goods. If these goods are generated from a recommender system, they will then directly affect the performance of the system. As such, our study attempts to enrich our understanding of the inducers and their effects in the recommendation performance. In addition, this paper provides a preliminary exploration of a mood-based filter to enhance the interaction between human and the system.