Methods and systems include receiving feedback from one of a plurality of users for one of a plurality of items, storing the received feedback and past feedback received from the plurality of users, wherein the received feedback and past feedback are stored as combinations of users and items having feedback given by the users, calculating relevance between the one of the plurality of items and another one of the plurality of items, based on first information indicating a degree of feedback given by the one of the plurality of users for the plurality of items, and second information indicating a degree of stored past feedback given by the plurality of users to the plurality of items, and immediately recommending items to the one of the plurality of users, the recommended items having the highest relevance to the one of the plurality of items, based on the calculated relevance.