Physica A 471 (2017) 147–153 Contents lists available at ScienceDirect Physica A journal homepage: www.elsevier.com/locate/physa Improving the recommender algorithms with the detected communities in bipartite networks Peng Zhang a,∗ , Duo Wang a , Jinghua Xiao a,b a School of Science, Beijing University of Posts and Telecommunications, Beijing 100876, PR China b State Key Laboratory of Information Photonics and Optical Communications, Beijing University of Posts and Telecommunications, Beijing 100876, PR China highlights • We noticed that the community structure can significantly affect the recommendations. • An improved recommender approach with the detected communities is proposed. • The results in both artificial and real networks show our approach performs better. • The inherent properties of networks should be considered in recommender systems. article info Article history: Received 1 August 2016 Available online 10 December 2016 Keywords: Recommender system Bipartite network Community structure abstract Recommender system offers a powerful tool to make information overload problem well solved and thus gains wide concerns of scholars and engineers. A key challenge is how to make recommendations more accurate and personalized. We notice that community structures widely exist in many real networks, which could significantly affect the recommendation results. By incorporating the information of detected communities in the recommendation algorithms, an improved recommendation approach for the networks with communities is proposed. The approach is examined in both artificial and real networks, the results show that the improvement on accuracy and diversity can be 20% and 7%, respectively. This reveals that it is beneficial to classify the nodes based on the inherent properties in recommender systems. © 2016 Elsevier B.V. All rights reserved. In the past decades of years, thanks to the rapid growth of the science and technology, our society is undergoing rapid transformation in almost all aspects. A number of society systems can be extracted as complex networks to research and varieties of information can be delivered on them, such as the traffic networks, the social networks, the Internet and so on. In recent years, the explosive increasing of the data on Internet not only can make our life more convenient, but also brings a serious information overload problem [1]: too much data to effectively extract the valuable ones that reduces the efficiency in making use of information. Some tools such as site navigation, search engines, professional database index are applied in filtering information. However, these approaches have their limits, particularly for personal requirements discovery. To solve this problem, recommender systems provides a tool to deliver the right information to the right individuals in the right occasions [2–5]. Recommender systems as an advanced intelligent platform serve for lots of web sites and provide personalized decisions. They are applied in a variety of applications. The most typical one which has outstanding prospects should be E-commerce ∗ Corresponding author. E-mail address: zhangpeng@bupt.edu.cn (P. Zhang). http://dx.doi.org/10.1016/j.physa.2016.11.076 0378-4371/© 2016 Elsevier B.V. All rights reserved.