摘要
在多媒体网络平台中,不仅社交网站允许用户自由发布资源和添加标签,越来越多的资源共享系统也开放给用户对资源、标签的组织管理权限。本文在分析了社会化标注系统的利弊后,采用推荐技术解决社会化标注系统中资源获取困难的问题,构建了基于社会化标注系统的个性化信息推荐模型,提出了从资源-标签-用户三个维度分别建立推荐组件,进而重组推荐资源集合实现对用户的个性化兴趣预测算法,并选取豆瓣网上的实例数据验证了算法的可行性和有效性。
On the Multi-media platform of network,apart horn the social websites allow users to release and add tags,more and more information resource sharing system open to all people permissions of organizing and managing items and tags.This paper analyzed the advantages and disadvantages,applying recommendation technology to solve the problem which is hard to obtain useful information.This paper constructed a personalized information recommendation model which is based on the folksonomy system.It proposed to design three recommendation components based on three dimentions of resource,tag and user respectively,then combined the three candidate sets in order to realize the user interests prediction.It took Douban Reading as an example to describe the proposed model and displayed the results of personalized information.
出处
《情报学报》
CSSCI
北大核心
2016年第5期549-560,共12页
Journal of the China Society for Scientific and Technical Information
基金
国家社会科学基金项目"大众分类中标签间语义关系挖掘研究"(批准号:12BTQ038)