With the rapid development of mobile Internet and the popularity of mobile devices,mobile content delivery has become one of the hottest topics in the pushing field. How to push content for users based on the behaviors of the mobile users has become the main task of mobile content push to improve its performance and user satisfaction. On the basis of analysis on traditional push technology and ad-vantages of automatic push technology as well as the dynamic characteristics of mobile data,a content push model based on the changes of mobile feature data has been presented,which can predict the user' s category label preference by SVM and select the relevant content from the content library for commendation to the users according to the change of user' s location and time. The experiments for its verifi-cation is conducted,which show that it is effective and feasible with good applicability.%随着移动互联网的快速发展以及移动设备的普及,移动信息内容推送已经成为当前的热点之一.根据移动终端用户的行为习惯确定内容推送,提高内容推送的性能和用户满意度,已成为移动内容推送的主要任务之一.为此,在概括分析传统推送技术及自动推送技术优势的基础上,根据移动数据的动态特征,提出了一种基于移动特征数据变化的内容推送模型.该模型根据用户所处位置和时间的变化,通过支持向量机(Support Vector Machine,SVM)算法来预测用户的类别标签偏好,并从内容库中选取相应的内容,推送给用户,并通过实验的方式对基于移动特征数据变化的内容推送模型的推送效果进行了验证.实验验证结果表明,所提出的模型有效、可行且适用性好.
展开▼
