文献类型： 外文期刊

作者： Xu, D. ¹ ; Hu, Y. X. ² ;

作者机构： 1.Tianjin Univ Technol & Educ, Sch Sci, Tianjin 300222, Peoples R China

2.Beijing Acad Agr & Forestry Sci, Inst Informat Agr Sci & Technol, Beijing 100097, Peoples R China

关键词： big data;message pushing;user characteristics;individuation

期刊名称：INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND ENVIRONMENTAL ENGINEERING (CSEE 2015)

ISSN：

年卷期： 2015 年

页码：

收录情况： SCI

摘要： People now are bothered by frequent and no specific pushing message via various APP. Big data provides data and technical support to enhance the quality of message pushing service. Big data analysis based on user characteristics can help message pushing to find the target users, so as to enhance the quality of pushing services. This paper proposes a mobile message pushing service platform based on user characteristics related big data. First of all, the platform architecture described in this paper intergrated vast amounts of user data via cloud services. Then, data-centric user characteristics model based on Map Reduce generated the user profile keyword index. At last, platform based on the degree of pushing message associated with the user, accurated positioning of the target population to avoid a large number of users being disturbed by unrelated pushing message. Experiment shows that it effectively improves the relevance and accuracy of message pushing.