Predicting the approximate time of event occurrence based on changes in the speed of sending messages in X social network

Document Type : Research Paper

Author

Faculty of Computer Engineering and IT, Payame Noor University, Tehran, Iran

Abstract

In recent years, the availability of virtual social network data and the mutual impact that real and virtual communities have on each other has led to much research in the field of virtual social network analysis. Detecting and predicting the occurrence of social events is one of the important applications of this field. In this paper, using a threshold structure, the approximate time of the event is predicted by analyzing the messages of social network X (former Twitter). In the proposed method, the data is first partitioned and preprocessed and then clustered using the distance-based Chinese restaurant process. Changes in the speed of sending messages to each cluster are used as an effective feature in predicting the approximate time of an event. Experiments conducted on almost 5 million tweets including 876 events show a prediction accuracy of 78%.

Keywords

[1] H. Becker, M. Naaman, and L. Gravano, Beyond trending topics: Real-world event identification on Twitter, Proc. Int. AAAI Conf. Web Soc. Media 5 (2011), no. 1, 438–441.
[2] P. Burnap, R. Gibson, L. Sloan, R. Southern, and M. Williams, 140 characters to victory?: Using Twitter to predict the UK 2015 general election, Electoral Stud. 41 (2016), 230–233.
[3] V. Dhar and E.A. Chang, Does chatter matter? The impact of user-generated content on music sales, J. Interact. Market. 23 (2009), no. 4, 300–307.
[4] W. Dou, X. Wang, W. Ribarsky, and M. Zhou, Event detection in social media data, IEEE VisWeek Workshop on Interactive Visual Text Analytics-Task Driven Analytics of Social Media Content, 2012, pp. 971–980.
[5] X. Fu, X. Jiang, Y. Qi, M. Xu, Y. Song, J. Zhang, and X. Wu, An event-centric prediction system for COVID-19, IEEE Int. Conf. Knowledge Graph (ICKG), 2020, pp. 195–202.
[6] Y.T. Huang and P.F. Pai, Using the least squares support vector regression to forecast movie sales with data from Twitter and movie databases, Symmetry 12 (2020), no. 4.
[7] J. Kalyanam, M. Quezada, B. Poblete, and G. Lanckriet, Prediction and characterization of high-activity events in social media triggered by real-world news, PloS one 11 (2016), no. 12, e0166694.
[8] A. J. McMinn, Y. Moshfeghi, and J.M. Jose, Building a large-scale corpus for evaluating event detection on Twitter, Proc. 22nd ACM Int. Conf. Inf. Knowledge Manag. 2013, pp. 409–418.
[9] N. Panagiotou, I. Katakis, and D. Gunopulos, Detecting events in online social networks: Definitions, trends and challenges, Solving Large Scale Learning Tasks, Challenges and Algorithms. (2016) 42–84.
[10] M. Thelwall, K. Buckley, G. Paltoglou, D. Cai, and A. Kappas, Sentiment strength detection in short informal text, J. Amer. Soc. Inf. Sci. Technol. 61 (2010), no. 12, 2544–2558.
[11] R. Vossen, Does chatter matter? Predicting music sales with social media, (2013), [online] Available at: https://www.basicthinking.de/blog/wpcontent/uploads/2013/06/Does-Chatter-Matter.pdf.
[12] S. Wasserman and J. Galaskiewicz, Advances in Social Network Analysis: Research in the Social and Behavioral Sciences, SAGE Publications, 1994.
[13] A. Yavari, H. Hassanpour, B. Rahimpour Cami, and M. Mahdavi, Event prediction in social network through Twitter messages analysis, Soc. Network Anal. Min. 12 (2022), no. 1.
[14] A. Yavari, H. Hassanpour, B. Rahimpour Cami, and M. Mahdavi, Election prediction based on sentiment analysis using Twitter data, Int. J. Engin. Trans. B: Appl. 35 (2022), no. 2, 372–379.
[15] A. Yavari and H. Hassanpour, Election prediction based on messages feature analysis in Twitter social network, Int. J. Engin. Trans. C: Aspects 36 (2023), no. 6, 1179–1184.
Volume 15, Issue 10
October 2024
Pages 157-162
  • Receive Date: 22 August 2023
  • Accept Date: 23 October 2023