摘要
针对忽视用户的社交关系变化可能得到不准确的推荐结果这一问题,为冷启动用户基于社交网络拓扑结构增量更新相似用户,并基于更新的相似用户给出准确的推荐.用户社交关系是动态变化的,然而现有基于社交网络的冷启动推荐却没有充分考虑社交关系的变更对推荐结果的影响.为了给冷启动用户实时准确的推荐,提出基于增量图形模式匹配的动态冷启动推荐方法(IGPMDCR),增量地更新冷启动用户的相似用户,为冷启动用户给出实时准确的推荐结果.在真实社交网站的数据集的实验结果表明,IGPMDCR可以在用户间社交关系变更的情况下,为冷启动用户给出实时准确的推荐结果.
Since ignoring the change of user＇s social relationships would lead to inaccurate recommendations,similar users for cold start users were updated based on social network topology incrementally,and accurate recommendations were provided based on updated similar users.User social relationships might be dynamically changed,however,the existing social network based cold-start recommendation methods did not fully consider the impact caused by the change of social relationships on recommendations as time pass by.In order to give accurate and timely in manner recommendations for cold start users,an incremental graph pattern matching based dynamic cold-start recommendation method（IGPMDCR）was proposed,which could update similar users for cold-start user incrementally,and give accurate and timely in manner recommendations for cold-start user.Experimental results on the real social network websites datasets show that IGPMDCR can give cold-start user accurately and timely in manner recommendations when user＇s social relationships are changing.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2017年第2期408-415,共8页
Journal of Zhejiang University：Engineering Science
基金
国家自然科学基金青年基金资助项目(61300115
61502410)
关键词
冷启动推荐
社交网络
增量图形模式匹配
社交网络拓扑结构
cold-start recommendation
social network
incremental graph pattern matching
topology of social network
作者简介
张亚楠（1981-）,男,讲师,博士,从事社交网络推荐等研究.ORCID：0000-0002-0633-826X.E-mail：ynzhang_1981@163.com
