Abstract
Studying the flow of influence in social media can allow insight into the nature of the agents involved and the corresponding actions that they take. In this paper, we study the influence of content among social media users with a concept called directed information (DI). Originally found in information theory, DI measures the amount of causal influence that an agent’s actions have on others. By estimating these quantities, influence networks are built that show the leaders and followers of a social circle. In order to demonstrate this technique, we extract tweets from the US presidential candidates and build an influence network. The time-varying influence is extracted using an extension of DI called adaptive directed influence (ADI), which is able to identify changes in influence over different timescales. Using the example of presidential candidates, we are able to show the power of building an influence network using DI and ADI and we compare and contrast with other relevant metrics.
This work was partially supported by ARO under grant #W911NF-12-1-0443.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Amblard, P.O., Michel, O.J.J.: The relation between granger causality and directed information theory: A review. Entropy 15, 113–143 (2013)
Chen, X., Hero, A.O., Savarese, S.: Multimodal video indexing and retrieval using directed information. IEEE Trans. Multimedia 14(1), 3–16 (2012)
Chen, X., Syed, Z., Hero, A.: EEG spatial decoding with shrinkage optimized directed information assessment. In: ICASSP 2012 Proceedings, pp. 577–580 (2012)
Jiao, J., Permuter, H.H., Zhao, L., Kim, Y.H., Weissman, T.: Universal estimation of directed information. IEEE Trans. Inf. Theor. 59(10), 6220–6242 (2013)
Massey, J.: Causality, feedback and directed information. In: Proceedings of 1990 International Symposium on Information Theory and its Applications, pp. 303–305 (1990)
Permuter, H.H., Kim, Y.H., Weissman, T.: Interpretations of directed information in portfolio theory, data compression, and hypothesis testing. IEEE Trans. Inf. Theor 57(6), 3248–3259 (2011)
Quinn, C., Kiyavash, N., Coleman, T.P.: Directed information graphs. IEEE Trans. Inf. Theor. 61(12), 6887–6909 (2015). coleman.ucsd.edu
Weissman, T., Kim, Y.H., Permuter, H.H.: Directed information, causal estimation, and communication in continuous time. IEEE Trans. Inf. Theor. 59(3), 1271–1287 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Oselio, B., Hero, A. (2016). Dynamic Directed Influence Networks: A Study of Campaigns on Twitter. In: Xu, K., Reitter, D., Lee, D., Osgood, N. (eds) Social, Cultural, and Behavioral Modeling. SBP-BRiMS 2016. Lecture Notes in Computer Science(), vol 9708. Springer, Cham. https://doi.org/10.1007/978-3-319-39931-7_15
Download citation
DOI: https://doi.org/10.1007/978-3-319-39931-7_15
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-39930-0
Online ISBN: 978-3-319-39931-7
eBook Packages: Computer ScienceComputer Science (R0)