SCIENTIA SINICA Informationis, https://doi.org/10.1360/SSI-2019-0274

A survey on knowledge graph based recommender systems

More info


Recommender system targets at providing accurate item recommendations to users with respect to their preferences, which has been widely applied to various online applications for addressing the problem of information explosion and enhancing user experiences. During the past decades, while tremendous efforts have been made in improving the performance of recommender systems, there still exist some long-standing challenges to address, including data sparsity, cold start, and result diversity. Along this line, an emerging research trend is to exploit the rich semantic information contained in the knowledge graph, which has been proven to be an effective way to enhance the capability of recommender systems. To this end, in this paper, we provide a focused survey on knowledge graph based reommender systems, through a holistic perspective of both technologies and applications. Specifically, we first briefly review the core concepts and classical algorithms of recommender system and knowledge graph. Then, we comprehensively introduce the representative and state-of-the-art works in this field, with respect to different strategies of exploiting knowledge graphs for recommender systems. Meanwhile, we also summarize some typical applications scenarios of knowledge graph based recommender system, for facilitating the hands-on practices of corresponding algorithms. Finally, we present our opinions on the prospects of knowledge graph based recommender system and suggest some future research directions in this area.

Copyright 2020 Science China Press Co., Ltd. 《中国科学》杂志社有限责任公司 版权所有