Recommender Systems: Basics, Methods and Applications

Abstract

Recommender systems efficiently deal with the information overload problems on World Wide Web. The recommender systems are software tools and techniques that recommend only the interesting and relevant items to the user by filtering the huge information on Internet. Thus, resulting in improved users’ satisfaction during navigation. The conventional, pure Web usage data based recommender systems are used efficiently. However, research from last few decades is focusing on improving the recommender systems’ performance by integrating domain knowledge along with the usage data. This paper discusses the recommender systems in detail. It explains the basic concepts of recommender systems and its types such as Collaborative Filtering (CF) based, Content based (CB), Hybrid, and Knowledge Based (KB) etc. Further CF based approach is discussed in detail. The paper also presents the major research problems in recommender systems, applications of recommender systems and the performance measures used for evaluation of recommender systems.