CoGrec: A Community-Oriented Group Recommendation Framework

• 作者: Bai Wang Bin Wu Jing Shi Xuelin Zeng Yu Liu Yunlei Zhang
• 来源期刊: 国际计算机前沿大会会议论文集
• 学科: 社会科学
• 关键词: Group RECOMMENDATION OVERLAPPING COMMUNITY Non-negative matrix FACTORIZATION
• 年，卷（期）: 2016,（1）
• 页码范围: 68-70
• 页数: 3页
• 分类号: C5

摘要

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.