全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于社会化媒体节点属性的信息预测

DOI: 10.13190/jbupt.201204.24.zhangch, PP. 24-27

Keywords: 社会化媒体,节点属性,预测模型,BP神经网络模型,票房预测

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对多数研究仅将社会化媒体作为数据来源的现状,深入分析社会化媒体特点,重点将节点属性分为静态和动态进行研究,提出基于预测目标的节点影响力的概念.在此基础上提出了一种基于节点属性进行信息预测的属性、节点数、倾向(ANV)模型.实验采用后向传播(BP)神经网络预测方法,通过新浪微博数据预测电影票房.仿真表明,带有节点属性的方法比没有节点属性的方法拟合和预测更为准确.

References

[1]  [1]Gruhl D, Guha R, Kumar R, et al. The predictive power of online chatter[C]//SIGKDD International Conference on Knowledge Discovery in Data Mining, 2005. New York: ACM Press, 2005: 78-87.
[2]  Liu Yang, Huang Xiangji, An Aijun, etal. A sentiment aware model for predicting sales performance using blog[C]//SIGIR, 2007. New York: ACM Press, 2007: 607-614.
[3]  Asur S, Huberman B A. Predicting the future with social media[C]// International Conference on Web Intelligence. IEEE Computer Society 2010. Washington: IEEE Press, 2010: 492-499.
[4]  Gilbert E, Karahalios K. Wide spread worry and the stock market[C]//ICWSM, 2010. Washington: AAAI Press, 2010: 59-65.
[5]  Bothos E, Apostolou D, Mentzas G. Using social media to predict future events with agent based markets[J]. Intelligent Systems, 2010, 25(6): 50-58.
[6]  Daniel C. Leveraging online social networks and external data sources to predict personality[C]//International Conference on Advances in Social Networks Analysis and Mining, 2011. Kaohsiung: IEEE Press, 2011: 428-433.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133