Computer Engineering and Applications ›› 2020, Vol. 56 ›› Issue (1): 1-10.DOI: 10.3778/j.issn.1002-8331.1907-0378

Previous Articles     Next Articles

Review of Social Recommender Systems

ZHANG Qishan, WENG Lijuan   

  1. School of Economics and Management, Fuzhou University, Fuzhou 350108, China
  • Online:2020-01-01 Published:2020-01-02

社会化推荐系统综述

张岐山,翁丽娟   

  1. 福州大学 经济与管理学院,福州 350108

Abstract: Recommender systems can help netizens find target information from?a large number of complex information, and can effectively improve netizens’ information retrieval ability, However, there are problems in data sparsity, cold-start and system performance in recommender systems. In order to solve these problems, some scholars proposed to apply social relations to the recommender systems, this method is an important way to improve the accuracy of recommendation, and has made important progress in many years of scientific research practice. Therefore, this research direction has increasingly become a field of concern for many scholars, and relevant research in this field is becoming more and more active. By sorting out the concept of social recommender systems and comparing with traditional recommender systems, the research status of social recommender systems is reviewed, hoping to find out new rules and seek new breakthroughs from the current research status, the prospects for future development of the social recommender systems are also discussed, in order to be helpful?to the later researchers.

Key words: recommender systems, social recommendation, collaborative filtering, matrix factorization, social media

摘要: 推荐系统可以帮助网民从大量纷繁的信息中找到目标信息,能有效提高网民信息检索能力,然而推荐系统存在数据稀疏性、冷启动以及系统性能方面的问题。为解决这方面的问题,提出将社交关系应用于推荐系统,该方法是提高推荐准确性的一个重要途径,在多年的科研实践中取得了重要进展,因此该研究方向也日益成为众多学者关注的领域,有关这方面的研究也越来越活跃。通过对社会化推荐系统概念进行梳理,对社会化推荐系统与传统推荐系统进行比较,回顾总结了社会化推荐系统的研究现状,希望能从研究现状中找出新规律,寻求新的突破点,并对社会化推荐系统的发展趋势进行展望,以期对后来研究者有所帮助。

关键词: 推荐系统, 社会化推荐, 协同过滤, 矩阵分解, 社交媒体