论文部分内容阅读
目前绝大多数推荐系统的用户偏好模型建立在注册用户和长期偏好基础上,而实际上用户常习惯于匿名访问信息检索系统。在总结高校数字图书馆用户偏好需求和信息源特性的基础上,本研究提出将匿名用户的短期偏好与各专业用户群的长期偏好相结合构建匿名用户偏好模型的方法,并利用向量空间技术实现模型设计,从而为高校数字图书馆推荐系统的用户偏好模型构建提供一种新思路。
At present, the user preference model of the vast majority of recommended systems is based on registered users and long-term preferences. In fact, users are often accustomed to anonymously accessing information retrieval systems. On the basis of summarizing the needs of users and the characteristics of information sources in university digital library, this study proposes to combine the short-term preferences of anonymous users with the long-term preferences of various professional user groups to build an anonymous user preference model and to use Vector Space Technology Model design, so as to provide a new idea for the building of user preference model of college digital library recommendation system.