社交网络中的消息流行度预测问题对于信息推荐和病毒式营销等应用具有重要意义。该文提出了一种基于传播模拟的消息流行度预测方法,首先使用最大熵模型学习并预测用户转发消息的概率,然后使用独立级联传播模型在真实的社会网络上模拟消息的传播过程,从而完成消息流行度的预测。该方法的优点在于更充分的利用了社会网络的结构和用户特征信息。该文在Twitter数据集上的实验结果表明,相对于基准方法,该文提出的方法具有更高的准确率和稳定性。
Abstract
Tweet popularity prediction in social network is very important for applications such as information recommendation and viral marketing. This paper proposes a new approach for tweet popularity prediction based on propagation simulation. The maximum entropy model is firstly used to learn the probabilities of users retweeting behaviors, and then the independent cascade model is used to simulate the diffusion processes of tweets in real social network. This approach benefits from using more information of social network structure and users. Experiments on Twitter dataset show that our approach is better in both precision and stability compared to baselines.
关键词
流行度预测 /
传播模型 /
最大熵模型
{{custom_keyword}} /
Key words
popularity prediction /
diffusion model /
maximum entropy model
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Wu F, Huberman B A. Novelty and collective attention [J]. National Academy of Sciences, 2007, 104(45): 17599.
[2] Lerman K, Hogg T. Using a model of social dynamics to predict popularity of news[C]//Proceedings of the ACM, USA, 2010: 621-630.
[3] Szabo G, Huberman B A. Predicting the popularity of online content [J]. Communications of the ACM, 2010, 53(8): 80-88.
[4] Bakshy E, Hofman J M, Mason W A, et al. Everyones an influencer: quantifying influence on twitter[C]//Proceedings of the ACM, HK, 2011: 65-74.
[5] Hong L, Dan O, Davison B D. Predicting popular messages in twitter[C]//Proceedings of the ACM, 2011: 57-58.
[6] Yang Z, Guo J, Cai K, et al. Understanding retweeting behaviors in social networks[C]//Proceedings of the 19th ACM CIKM. Canada, 2010: 1633-1636.
[7] Galuba W, Aberer K, Chakraborty D, et al. Outtweeting the twitterers-predicting information cascades in microblogs[C]//Proceedings of the 3rd conference on Online social networks. USA:USENIX Association, 2010: 3-3.
[8] Suh B, Hong L, Pirolli P, et al. Want to be retweeted? Large scale analytics on factors impacting retweet in Twitter network[C]//Proceedings of the SocialCom. USA:IEEE, 2010: 177-184.
[9] Kempe D, Kleinberg J, Tardos Maximizing the spread of influence through a social network[C]//Proceedings of the ninth ACM SIGKDD. USA, 2003: 137-146.
[10] Ver S G, Ghosh R, Lerman K. What stops social epidemics[C]//Proceedings of the Fifth International AAAI Conference on Weblogs and Social Media. USA, 2011: 377-384.
[11] Berger A L, Pietra V J D, Pietra S A D. A maximum entropy approach to natural language processing [J]. Computational linguistics, 1996, 22(1): 39-71.
[12] Salganik M J, Dodds P S, Watts D J. Experimental study of inequality and unpredictability in an artificial cultural market [J]. Science, 2006, 311(5762): 854-856.
[13] Kwak H, Lee C, Park H, et al. What is Twitter, a social network or a news media?[C]//Proceedings of the 19th ACM, USA, 2010: 591-600.
[14] Wu S, Hofman J M, Mason W A, et al. Who says what to whom on twitter[C]//Proceedings of the 20th ACM, USA, 2011: 705-714.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(61202213,61203298,60933005,61173008,61003166);国家“973”重点基础研究发展计划项目基金(2012CB316303)
{{custom_fund}}
PDF(1569 KB)
