微博中基于用户偏好的信息传播研究

发布时间：2018-04-26 21:06

本文选题：社会网络分析 + 偏好建模　；参考：《计算机工程与科学》2013年03期

【摘要】：对用户行为的分析是Web社会网络分析中的重要课题之一,包括对用户个体和群体的行为和偏好进行建模。为预测Web社会网络中的信息传播趋势,以微博为平台,基于对用户历史行为信息的统计分析,抽取出影响用户决策的因素,进而定性地分析各因素之间的关联关系,从而挖掘出用户潜在的偏好。在此基础上,利用CP-nets偏好表达工具建立了用户偏好模型,直观有效地表达各因素之间的偏好关系,同时可根据其导出图输出最优的特性来选取最佳特征项组合,以便用户进行转发决策度量。
[Abstract]:The analysis of user behavior is one of the most important topics in the analysis of Web social network, including the modeling of individual and group behavior and preference of users. In order to predict the trend of information dissemination in Web social network, based on Weibo, based on the statistical analysis of user's historical behavior information, the factors that affect the user's decision are extracted, and the relationship among the factors is analyzed qualitatively. In order to discover the potential preferences of users. On this basis, the user preference model is established by using the CP-nets preference expression tool, which can express the preference relationship between the factors directly and effectively. At the same time, the optimal feature combination can be selected according to the output optimal characteristics of the derived graph. So that the user can measure the forwarding decision.
【作者单位】：烟台大学计算机科学与技术学院智能信息处理实验室;
【基金】：国家自然科学基金资助项目(61170224) 山东省自然科学基金资助项目(ZR2011FL018,ZR2012FL07) 山东科技发展计划资助项目(2012GGB01017) 山东省教育厅科技计划资助项目(J11LG35,J10LG27)
【分类号】：TP393.09;TP311.13

【参考文献】