Abstract—Recommender systems are being used extensively to assist users in making right decisions in this present generation of information overload. Due to continuous exponential increase of online information and data, recommender systems are very much challenged by the issue of discovering the relevant information from this pool. As an effort to address this problem, research has been conducted to improve the recommendation quality of recommender systems. However more focus has been on improving recommendation accuracy while aggregate recommendation quality received less attention. In order to ensure that the recommendations are more useful to users, diversity has to be factored in. This will ensure that users are recommended items that they would have not been able to discover by themselves. This paper reviews some of the techniques employed to ensure aggregate diversity in recommended items.

Index Terms—collaborative filtering, cross-check approach, recommendation diversity, recommender systems, ranking functions

Cite: Sebabatso J. Metla, Tranos Zuva, and Seleman M. Ngwira, "Aggregate Diversity Techniques in Recommender Systems," Lecture Notes on Information Theory, Vol. 2, No. 3, pp. 238-242, September 2014. doi: 10.12720/lnit.2.3.238-242