This disclosure is directed to computational, closed-loop user feedback systems and methods for ranking or updating beliefs for a user based on user feedback. The systems and methods are based on a data-agnostic user feedback formulation that uses user feedback to automatically rank beliefs for a user or update the beliefs. The methods and systems are based on a general statistical inference model, which, in turn, is based on an assumption of convergence in user opinion. The closed-loop user feedback methods and systems may be used to rank or update beliefs prior to inputting the beliefs to a recommender engine. As a result, the recommender engine is expected to be more responsive to customer environments and efficient at deployment and reducing the level of unnecessary user recommendations.