EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Predicting the Popularity of Information on Social Platforms without Underlying Network Structure

Leilei Wu, Lingling Yi, Xiao-Long Ren iD, Linyuan Lü

Entropy · 09.06.2023 · 25 (6) · 916

DOI: 10.3390/e25060916 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 09.06.2023. Verfasst von Leilei Wu und 3 weiteren Autor:innen. Beteiligte Einrichtung: Yangtze Delta Region Institute (Huzhou), University of Electronic Science and Technology of China, Huzhou 313001, China. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: STI 2030-Major Projects, China Postdoctoral Science Foundation, Sichuan Science and Technology Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The ability to predict the size of information cascades in online social networks is crucial for various applications, including decision-making and viral marketing. However, traditional methods either rely on complicated time-varying features that are challenging to extract from multilingual and cross-platform content, or on network structures and properties that are often difficult to obtain. To address these issues, we conducted empirical research using data from two well-known social networking platforms, WeChat and Weibo. Our findings suggest that the information-cascading process is best described as an activate–decay dynamic process. Building on these insights, we developed an activate–decay (AD)-based algorithm that can accurately predict the long-term popularity of online content based solely on its early repost amount. We tested our algorithm using data from WeChat and Weibo, demonstrating that we could fit the evolution trend of content propagation and predict the longer-term dynamics of message forwarding from earlier data. We also discovered a close correlation between the peak forwarding amount of information and the total amount of dissemination. Finding the peak of the amount of information dissemination can significantly improve the prediction accuracy of our model. Our method also outperformed existing baseline methods for predicting the popularity of information.

Schlagwörter

Stichwörter der Autor:innen

cascade predictioninformation cascadeinformation diffusiononline social networkpopularity prediction

Autor:innen und Institutionen

  • Yangtze Delta Region Institute (Huzhou), University of Electronic Science and Technology of China, Huzhou 313001, China.
  • Department of Physics, University of Fribourg, CH-1700 Fribourg, Switzerland.
  • Alibaba Business School, Hangzhou Normal University, Hangzhou 311121, China.
  • Tencent Technology (Shenzhen) Co., Ltd., Shenzhen 518000, China.
  • Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu 611731, China.

ORCID-Profile: Xiao-Long Ren

Förderung

  • STI 2030-Major Projects
  • China Postdoctoral Science Foundation
  • Sichuan Science and Technology Program
  • Project of Huzhou Science and Technology Bureau
  • UESTCYDRI research start-up
  • STI 2030–Major Projects
  • New Cornerstone Science Foundation

Zitieren

APA

Wu, L., Yi, L., Ren, X. L., & Lü, L. (2023). Predicting the Popularity of Information on Social Platforms without Underlying Network Structure. Entropy, 25(6), 916. https://doi.org/10.3390/e25060916

Vancouver

Wu L, Yi L, Ren XL, Lü L. Predicting the Popularity of Information on Social Platforms without Underlying Network Structure. Entropy. 2023;25(6):916. doi:10.3390/e25060916

BibTeX
@article{Wu2023,
  author = {Wu, Leilei and Yi, Lingling and Ren, Xiao-Long and Lü, Linyuan},
  title = {Predicting the Popularity of Information on Social Platforms without Underlying Network Structure},
  journal = {Entropy},
  year = {2023},
  volume = {25},
  number = {6},
  pages = {916},
  doi = {10.3390/e25060916}
}
Kostenlos starten →