Prediction of retweet counts by a back propagation neural network

DENG Qing; MA Yefeng; LIU Yi; ZHANG Hui

doi:10.16511/j.cnki.qhdxxb.2015.24.012

PDF(1021 KB)

Journal of Tsinghua University(Science and Technology) ›› 2015, Vol. 55 ›› Issue (12) : 1342-1347. DOI: 10.16511/j.cnki.qhdxxb.2015.24.012

PHYSICS AND ENGINEERING PHYSICS

Prediction of retweet counts by a back propagation neural network

{{article.zuoZhe_EN}}

Author information +

History +

Abstract

Twitter has become a major platform for expressing and gathering information to change people's opinions and lives. Retweets are a key mechanism for information diffusion. The retweet mechanism can be a useful method to guide public opinion and contribute to emergency responses. This paper considers a case study of the conflicts between urban management officials (known as Chengguan in China) and the public. This study focused on factor analysis and prediction of a tweet's popularity based on a back propagation (BP) neural network during a crisis. The weighted analysis of various factors from the perspectives of the posters and the content of the microblog messages shows how some factors, including the user's activity, hashtag, visual information, mentioning others and posting time, influences a message's popularity. The results show that followers are more attracted by a tweet's content rather than its poster. The prediction problem is changed into a pattern classification problem to predict the retweet count using a back propagation (BP) neural network. The stability of the results was tested by changing the number of samples.

Key words

Twitter / retweets / back propagation (BP) neural network / prediction / factors / weighted analysis / emergency responses

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

DENG Qing, MA Yefeng, LIU Yi, ZHANG Hui. Prediction of retweet counts by a back propagation neural network[J]. Journal of Tsinghua University(Science and Technology). 2015, 55(12): 1342-1347 https://doi.org/10.16511/j.cnki.qhdxxb.2015.24.012

References

[1] GUAN Wanqiu, GAO Haoyu, YANG Mingmin. Analyzing user behavior of the micro-blogging website Sina Weibo during hot social events [J]. Physica A, 2014, 395: 340-351.
[2] Rudat A, Buder J, Hesse F W. Audience design in Twitter: Retweeting behavior between informational value and followers' interests [J]. Computers in Human Behavior, 2014, 35: 132-139.
[3] Morchid M, Dufour R, Bousquet P M, et al. Feature selection using principal component analysis for massive retweet detection [J]. Pattern Recognition Letters, 2014, 49: 33-39.
[4] 吴凯, 季新生, 刘彩霞. 基于行为预测的微博网络信息传播建模 [J]. 计算机应用研究, 2013, 30(6): 1809-1813. WU Kai, JI Xinsheng, LIU Caixia. Modeling information diffusion based on behavior predicting in microblog [J]. Application Research of Computers, 2013, 30(6): 1809-1813. (in Chinese)
[5] LIANG Bin, LIU Yiqun, ZHANG Min, et al. Searching for people to follow in social networks [J]. Expert Systems with Applications, 2014, 41(16): 7455-7465.
[6] Armentano M G, Godoy D, Amandi A A. Followee recommendation based on text analysis of micro-blogging activity [J]. Information Systems, 2013, 38(8): 1116-1127.
[7] PetroviS, Osborne M, Lavrenko V. RT to Win! Predicting Message Propagation in Twitter [C] // Fifth International AAAI Conference on Weblogs and Social Media (ICWSM). Barcelona, Spain: Association for the Advancement of Artificial Intelligence, 2011.
[8] 张旸, 路荣, 杨青. 微博客中转发行为的预测研究 [J]. 中文信息学报, 2012, 26(4): 109-115. ZHANG Yang, LU Rong, YANG Qing. Prediction retweeting in Microblogs [J]. Journal of Chinese Information Process, 2012, 26(4): 109-115. (in Chinese)
[9] Suh B, Hong L, Pirolli P, et al. Want to be retweeted? Large scale analytics on factors impacting retweet in twitter network [C] // 2010 IEEE Second International Conference on Social Computing (SocialCom). Minneapolis, USA: IEEE, 2010.
[10] 李英乐, 于洪涛, 刘力雄. 基于SVM的微博转发规模预测方法 [J]. 计算机应用研究, 2013: 30(9), 2594-2597. LI Yingle, YU Hongtao, LIU lixiong. Predict algorithm of micro-blog retweet scale based on SVM [J]. Application Research of Computers, 2013, 30(9): 2594-2597. (in Chinese)
[11] ZHANG Yiwen, QI Jiayin, FANG Binxing, et al. The indicator system based on BP neural network model for net-mediated public opinion on unexpected emergency [J]. China Communications, 2011, 8(2): 42-51.
[12] 何长虹, 黄全义, 申世飞, 等. 基于BP神经网络的森林可燃物负荷量估测 [J]. 清华大学学报: 自然科学版, 2011, 51(2): 230-233. HE Changhong, HUANG Quanyi, SHEN Shifei, et al. Forest fuel loading estimates based on a back propagation neutral network [J]. Journal of Tsinghua University: Science and Technology, 2011, 51(2): 230-233. (in Chinese)
[13] 杨淑娥, 黄礼. 基于BP神经网络的上市公司财务预警模型 [J]. 系统工程理论与实践, 2005, 1(1): 12-18. YANG Shu'e, HUANG Li. Financial crisis warning model based on BP neural network [J]. System Engineering-Theory & Practice, 2005, 1(1): 12-18. (in Chinese)
[14] 张天云, 陈奎, 魏伟, 等. BP神经网络法确定工程材料评价指标的权重 [J]. 材料导报, 2012, 26(2): 159-163. ZHANG Tianyun, CHEN Kui, WEI Wei, et al. The determination of index weights for comprehensive evaluation engineering materials with BP neural network [J]. Materials Review, 2012, 26(2): 159-163. (in Chinese)
[15] 刘敏, 孙树栋. 基于ANN的电子商务水平测度指标权重的确定方法 [J]. 电子商务, 2006, 4(88): 136-140. LIU Min, SUN Shudong. Research on method of computing indicator weight for E-commerce development level estimaion based on ANN [J]. Journal of E-commerce, 2006, 4(88): 136-140. (in Chinese)
[16] WANG Zhiliang, LI Yongchi, Shen R F. Correction of soil parameters in calculation of embankment settlement using a BP network back-analysis model [J]. Engineering Geology, 2007, 91(2): 168-177.
[17] REN Chao, AN Ning, WANG Jianzhou, et al. Optimal parameters selection for BP neural network based on particle swarm optimization: A case study of wind speed forecasting [J]. Knowledge-Based Systems, 2014, 56: 226-239.