Skip to main content



Diffusion of Information in Social Media Networks

This paper addresses the idea of diffusion of information in social media networks. Typical studies would follow an approach of seeing how a specific network can be affected by looking at each node and its effect on other nodes and how this affect may cascade through the network. This paper looks to develop a model that does not depend on literal knowledge of the structure of the network and instead is able to determine the influence of one node based on the rate of diffusion of information through an implicit network. The model was validated using a set of several tweets, over 500 million, and using several news articles and blog posts. The researchers found the Linear Influence Model to accurately predict the dynamics of information diffusion in these networks.

From the use of this model many things were discovered. For example, it was discovered that one limitation of the model is that there may be certain individuals who differ significantly from the typical node and thus their influence cannot be accurately predicted with the model using the rate of diffusion. Instead, these individuals must be modeled separately for a more accurate representation of the diffusion network. It was found that the model also changes based on the type of website and topic of information as these parameters are very important. One discovery was that some short news phrases were mainly influenced by large media websites while hashtags on Twitter were influenced by a much larger set of nodes, twitter users.

This study relates back to the discussion of diffusion networks we have had in class and looks to see a new way of modeling the networks that is not strictly based on knowledge of a known network. Here, researchers are able to somewhat accurately model news media and social media by predicting the influence one node may have on the overall network.

https://cs.stanford.edu/~jure/pubs/lim-icdm10.pdf

Comments

Leave a Reply

Blogging Calendar

November 2018
M T W T F S S
 1234
567891011
12131415161718
19202122232425
2627282930  

Archives