Research on Redundancy of Implicit Feedback Information in Web Browsing
丁丁 魏 · Computer Science and Application · 2017
针对用户网络浏览过程中的隐式反馈信息数据量大但偏好信息表达不明确的问题，提出了基于多元尺度分析理论的网络浏览隐式反馈信息冗余性分析方法。该方法将用户对不同网站浏览的隐式反馈信息作为时间序列，分别计算6个静态特征，通过建立隐式反馈行为特征矩阵，计算不相似度矩阵，从而实现特征矩阵在低维空间的重构，以展现用户不同网络浏览行为的冗余性，为后期将研究结论应用于推荐系统奠定基础。实验结果表明，该方法可以有效地分析网络浏览隐式反馈信息的冗余性，得到具有指导性的隐式反馈信息选取原则。 Aiming at the problem that the amount of implicit feedback information in the user's web browsing process is large but the information is not clear, a redundancy analysis method of web-browsing implicit feedback information based on multi-dimensional aspect analysis was proposed. Taking the user’s implicit feedback information on different website as time series, 6 static characteristics have been calculated. And by constructing the implicit feedback behavior feature matrix, the non-similarity matrix is calculated to reconstruct the feature matrix in the low-dimensional space. It can show the redundancy of user's different web browsing behavior, which will lay the foundation for applying the research conclusion to the recommended system. The results show that the proposed method can effectively analyze the redundancy of the implicit feedback information in web browsing, and obtain the guiding principle of implicit feedback information’s selection.