A self-learning collaborative filtering algorithm in recommendation system
Anyu Dai, Yangyang Xia, Yue Gui · 2017
Recommender systems have made great progress over the past few years and been widely used in e-commerce. Collaborative filtering is the most common method in personalized recommendation systems. Many improvements have been made in order to improve the accuracy of the recommendations, but most of them are too complex and not versatile. Because RMSE is an important measure of recommended accuracy, this paper uses RMSE as the objective function to construct a function on the basis of the original model, so that RMSE is a convex function, which can be optimized by the gradient method. Experiments on a movie dataset show that the method has a good performance, and the method can be combined with a number of different methods of similarity, all of them have a better effect.