Models Andmethods for Forecasting Recommendations for Collaborative Recommender Systems
This article analyzes the current state of models and methods for constructing recommender systems. The main classes of tasks that solve recommender systems are highlighted. The features of the application of the method of collaborative (joint) filtering are shown. A mixed numerical-categorical clustering method for searching for user groups that uses numerical rating and demographic characteristics of users has been developed, a hybrid method for searching for user groups has been developed that uses the coefficient of usersubject matrix sparseness.