基于用户偏好行为数据的数字图书馆资源推荐方法研究

Research on Resource Recommendation Method of Digital Library Based on User Preference Behavior Data

为给读者提供更加精准的推荐服务,提出了一种基于用户偏好行为数据的数字图书馆资源推荐方法。从用户对数字图书馆资源的偏好行为中提取浏览、评论、下载三种数据,结合用户资源评分,构建用户偏好行为数据推荐模型。在确定模型中参数取值以确保模型方法最优推荐精度后,与传统推荐方法进行实验比较。通过实验验证该方法模型在准确率、召回率方面均优于传统的协同过滤方法,有效解决了资源评分数据稀疏性的问题。

In order to provide readers with more accurate data recommendation services,a resource recommendation method of digital library based on user preference behavior data is proposed.Three kinds of data are extracted from users’preference behavior for digital library resources:Browsing,commenting and downloading.Combining the user preference behavior data score with the user resource score,the user preference behavior data recommendation model is constructed.After determining the parameter values in the model to ensure the optimal recommendation accuracy of the model method,the experiment is compared with the collaborative filtering method.The experimental results show that the proposed model is superior to the traditional collaborative filtering method in terms of accuracy and recall rate,and effectively solves the problem of sparsity of resource rating data.