Recommender systems aim to support decision-makers by providing decision advice. We offer multi-criteria decision recommendations based on a performance matrix and a partial order on criteria submitted by the user. Our method is to aggregate performance measures over all criteria based on inferences about preferences from the decision-maker’s input. After reviewing some multicriteria aggregation operators, we present a recommender system that uses the Choquet integral of a fuzzy measure to determine a total ordering of the alternatives.