Being accurate is not enough: measuring and optimizing the diversity of recommendations

Preprint
Publisher Website
Google Scholar
  • 20 August 2008
Abstract
Created to handle the ever increasing amount of information, recommender systems represent a prominent challenge of information science. In the case when users only collect interesting objects (no ratings allowed), we introduce several quantities measuring both recommendation accuracy and diversity. We use them to obtain a systematic comparison of the available recommendation methods. In addition, we propose a hybrid method and show that it outperforms the original methods.

This publication has 0 references indexed in Scilit: