METHOD AND APPARATUS FOR RECOMMENDING ITEMS BASED ON DEEP LEARNING

摘要

Disclosed is a method for recommending items based on deep learning. A method of recommending an item by using initial evaluation information, which is information including an evaluation result value in which each of a plurality of users evaluates at least one of a plurality of items according to an embodiment of the present invention, allows the user to Generating predetermined evaluation information by inputting a predetermined initial result value as an evaluation result value corresponding to an unevaluated item which is an item not evaluated; Learning a deep learning model based on a predetermined differential weight applied to the update evaluation information and the error corresponding to the unevaluated item in the update evaluation information; Generating predictive evaluation information predicting a result of evaluating the plurality of items by the plurality of users based on the deep learning model and the initial evaluation information; And recommending a plurality of recommended items among the plurality of items to a recommendation target that is one of the plurality of users using the prediction evaluation information.