266_融合热点话题的微博转发预测研究.pdfVIP

融合热点话题的微博转发预测研究 1 * 2,3,4 1 2 陈江 ，刘玮 ，巢文涵 ，王丽宏 (1.北京航空航天大学，北京市，100191； 2.国家计算机网络应急技术处理协调中心，北京市，100029 ； 3.中国科学院计算技术研究所，北京 100190； 4. 中国科学院大学，北京 100049) 摘要：微博转发行为是实现信息传播的重要方式，微博转发预测对微博影响力分析、微博话题分析具有重 要价值。现有微博转发预测研究大多围绕消息属性、用户属性等微博自身特征，本文提出融合热点话题的 微博转发预测方法，对背景热点话题内容和传播趋势对用户转发行为的影响进行量化分析，提出融合背景 热点信息的转发兴趣、转发活跃度、行为模式等特征，并基于分类算法建立了面向热点话题相关微博的转 发预测模型，在真实数据上的实验结果表明，本文方法的预测准确性达到96.6%，提升幅度最高达到12.14%。 关键词：转发行为；转发预测；热点话题 中图分类号：TP391 文献标识码：A Research on Microblog Forwarding Prediction based on Hot Topics Jiang Chen1, Wei Liu* 2,3,4, Wenhan Chao1, Lihong Wang2 (1.Beihang University, Beijing 100191; 2. National Computer network Emergency Response technical Team/Coordination Center of China, Beijing 100029; 3. Institute of Computing Technology, Chinese Academy of Sciences, Beijing, 100190; 4. University of Chinese Academy of Sciences, Beijing, 100049) Abstract: Microblog forwarding is an important way to achieve the information dissemination, Microblog forwarding prediction is of great value in the analysis of microblog influence and microblog topic analysis. Existing microblog forward prediction methods mostly focus on microblog attributes and user attributes. In this paper, a microblog forwarding prediction method based on hot topics is proposed. We quantitatively analyzed the impact of hot topics content and transmission tendency on users’ forwarding behavior, and then introduced features that fused with hot topics such as forwarding interest, forwarding activity and behavior pattern. Finally, we established the hot topic oriented microblog forwarding prediction mode