面向“国图公开课”的数字图书馆个性化信息推荐服务研究

Research on Personalized Information Recommendation Service of Digital Library for National Library Open Course

面向“国图公开课”,构建联系用户兴趣与长尾资源的个性化信息服务模型。结合“国图公开课”数据资源池与用户行为池的业务场景,挖掘用户隐式兴趣信息,深化数字图书馆个性化信息推荐服务模式。基于用户显式行为的兴趣建模不能够良好地表征用户的个性化知识需求,挖掘用户的隐式兴趣特征并融入到个性化信息推荐服务中,以此改善用户兴趣矩阵数据稀疏引起的图书馆信息服务性能不理想状况,有效提升现代数字图书馆的用户体验及资源利用效能。

For the^National Library Open Course",the authors construct a personalized information service model that links user interests with long tail resources.By combining the business scenarios of the*'National Library Open Course"data resource pool and the user behavior pool,the user implicit interest information mining is introduced to deepen the digital library personalized information recommendation service mode.The interest modeling based on user explicit behavior cannot well represent the user’s personalized knowledge needs,so the user's implicit interest characteristics are mined and integrated into the personalized information recommendation service,so as to improve the poor performance of the library information service caused by the sparse user interest matrix data,and effectively improving the user experience and resource utilization efficiency of modern digital libraries.