Modern Approaches to Building Recommender Systems for Online Stores

Abstract

The article presents current approaches to solving the problem of building recommender systems designed to intellectualize the user interface of online stores. Much attention is paid to modern methods of building recommender systems, analysing their strengths and weaknesses. Of greatest interest are the criteria for selecting effective methods for specific online stores and the authors’ concept of a typical recommender system of electronic commerce.

Keywords

• collaborative filtering
• combined co-filtering
• content filtering
• typical recommender system