Recently, group recommendation becomes substantially significant when it frequently happens that a group of users need to determine which item (e.g. movie, music, restaurant, etc.) to choose. In this paper we employ the information of friend network to propose a Community-Oriented Group Recommendation framework (CoGrec) consisting of non-negative matrix factorization based user profile generation, community detection based group identification, and overlapping community membership based group decision. Along with four inherent aggregation and allocation strategies, our proposed framework is evaluated through extensive experiments on real-world datasets. The experimental results show that the proposed framework is promising and more accurate when the given friend network is much denser, which is suitable for modern review and rating systems.
CITATION STYLE
Liu, Y., Wang, B., Wu, B., Zeng, X., Shi, J., & Zhang, Y. (2016). CoGrec: A community-oriented group recommendation framework. In Communications in Computer and Information Science (Vol. 623, pp. 258–271). Springer Verlag. https://doi.org/10.1007/978-981-10-2053-7_24
Mendeley helps you to discover research relevant for your work.