摘要
近年来,在线社交网络成为人们工作、生活不可或缺的信息共享与交流工具,如何对海量庞杂、大范围时空关联的用户行为信息进行认知并据此提供个性化的推荐服务,已成为在线社交网络发展重点关注的问题。为此,提出了一种基于用户行为认知的在线社交网络协同推荐框架,在对用户特征、文本信息及兴趣偏好等行为进行认知的基础上,利用协同过滤算法,实现个性化的推荐服务。实验结果验证了提出的基于用户行为认知的协同推荐策略具有较好的稳定性和实际应用效果。
Recently, online social network （OSN） has become essential tools for information sharing and communication in people＇s work and life. How to cognitive massive, complex, large-area and spatiotemporal association user behavior information and provide personalized relcommendation services have become problems need special attention in development of OSN. Thus, a frame of collaborative recommendation for OSN based on user behavior cognitive was proposed which used collaborative filtering algorithm to provide personalized recommendation services on the basis of analysis of behavior of user characteristics, text information and interest preferences, etc. Experimental results verify that the proposed collaborative recommendation strategy based on user behavior cognitive has good stability and actual application effect.
出处
《电信科学》
北大核心
2015年第10期108-114,共7页
Telecommunications Science
基金
江苏省产学研联合创新资金资助项目(No.BY2012035)
南京邮电大学人才引进项目(No.NY213050
No.NY214098)~~
关键词
在线社交网络
行为认知
协同推荐
用户兴趣模型
online social network, behavior cognitive, collaborative recommendation, user interest model
