AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
Experimental results demonstrate that our method significantly outperforms the state-of-the-art method which neglects the structural characteristics of social networks

Popularity prediction in microblogging network: a case study on sina weibo

Proceedings of the 22nd international conference on World Wide Web companion, no. null (2013): 177-178

Cited: 158|Views108
EI WOS SCOPUS

Abstract

Predicting the popularity of content is important for both the host and users of social media sites. The challenge of this problem comes from the inequality of the popularity of content. Existing methods for popularity prediction are mainly based on the quality of content, the interface of social media site to highlight contents, and the ...More

Code:

Data:

0
Introduction
  • Popularity prediction on social networks can help users sift through the vast stream of online contents and enable advertisers to maximize revenue through differential pricing for access to content or advertisement placement.
  • Popularity prediction is challenging since numerous factors can affect the popularity of online content.
  • Popularity is very asymmetric and broadly-distributed.
  • Several pioneering work devoted to the characteristics and mechanisms of information diffusion [1, 2, 3].
  • Several efforts have been made to study the popularity prediction on social networks.
Highlights
  • Popularity prediction on social networks can help users sift through the vast stream of online contents and enable advertisers to maximize revenue through differential pricing for access to content or advertisement placement
  • Experimental results demonstrate that our method significantly outperforms the state-of-the-art method which neglects the structural characteristics of social networks
  • Encouraged by the work in [7], we investigate whether the final popularity of a tweet is well indicated by the structural characteristics of the network consisting of users that re-tweet the tweet at an earlier time
  • We have studied how to predict the popularity of short message in Sina Weibo
  • We find that structural characteristics provide strong evidence for the final popularity
  • Based on such a finding, we propose two approaches by incorporating the early popularity with the link density and the diffusion depth of early adopters
Methods
  • The authors use Sina Weibo dataset published by WISE 2012 Challenge1.
  • The data set consists of 16.6 million tweets.
  • This data set contains a snapshot of the social network of Sina Weibo.
  • The predictions are evaluated with RMSE and MAE.
  • As reported in Table 1, the approach incorporating the link density significantly reduces the prediction error compared with the baseline, and the approach incorporating the diffusion depth performs even
Results
  • The authors first study the structural characteristics of the forward path of tweets.
  • Encouraged by the work in [7], the authors investigate whether the final popularity of a tweet is well indicated by the structural characteristics of the network consisting of users that re-tweet the tweet at an earlier time.
  • The first measurement is link density.
  • Among all users that have forwarded the tweet k at time ti, link density is the ratio of the number of followship links to the number of all possible links.
  • The other measurement is the diffusion depth, which is the longest length of the path from the submitter to any user that has retweeted the tweet k at time ti
Conclusion
  • The authors have studied how to predict the popularity of short message in Sina Weibo.
  • A low link density and a deep diffusion usually lead to wide spreading, capturing the intuition that a diverse group of individuals spread a message to wider audience than a dense group.
  • Based on such a finding, the authors propose two approaches by incorporating the early popularity with the link density and the diffusion depth of early adopters.
  • The authors' finding provides a new perspective to understand the popularity prediction problem and is helpful to build accurate prediction models in the future
Tables
  • Table1: Prediction error of three approaches
Download tables as Excel
Funding
  • This work is funded by the National Natural Scientific Foundation of China under grant Nos. 61232010, 61202215 and National Basic Research Program of China (the 973 program) under grant No 2013CB329602
  • This work is partly funded by the Beijing Natural Scientific Foundation of China under grant No 4122077
  • This work is also supported by Key Lab of Information Network Security, Ministry of Public Security
Reference
  • A.-L. Barabasi. The origin of bursts and heavy tails in human dynamics. Nature, 435(7039):207–211, 2005.
    Google ScholarLocate open access versionFindings
  • F. Wu, B. Huberman. Novelty and collective attention. Proc. Natl. Acad. Sci., 104(45):17599-17601, 2007.
    Google ScholarLocate open access versionFindings
  • J. Yang, J. Leskovec. Patterns of temporal variation in online media. In Proc. of WSDM’11, pages 177–186, Feb. 2011, Hong Kong.
    Google ScholarLocate open access versionFindings
  • G. Szabo, B. A. Huberman. Predicting the popularity of online content. Commun. ACM, 53(8):80–88, 2010.
    Google ScholarLocate open access versionFindings
  • K. Lerman, T. Hogg. Using a model of social dynamics to predict popularity of news. In Proc. of WWW ’10, pages 621–630, Apr. 2010, Raleigh, USA.
    Google ScholarLocate open access versionFindings
  • L. Hong, O. Dan, B. D. Davison. Predicting popular messages in twitter. In Proc. of WWW’11, pages 57–58, Mar. 2011, Byderabad, India.
    Google ScholarLocate open access versionFindings
  • J. Ugander, L. Backstrom, C. Marlow, J. Kleinberg. Structural diversity in social contagion. Proc. Natl Acad. Sci., 109(14):5962–5966, 2012.
    Google ScholarLocate open access versionFindings
0
Your rating :

No Ratings

Tags
Comments
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn