Initial indicators of topic success in Twitter: Using topology entropy to predict the success of Twitter hashtags

The recent and dramatic increase in social media use by the general population across the globe has proven to be a valuable resource for understanding social dynamics. In this paper we focus on metrics that provide early indicators of the eventual impact of events, and attempt to show correlations between these early indicators and real world events. Specifically, a measure of early-stage diffusion between social network communities is examined as a predictor of the eventual effect of a given meme. Online social media dynamics are examined in Twitter where we tracked hashtags related to the 2012 US elections and the now-infamous campaign by the European Commission to encourage female interest in science. Community Entropya measure of topological information spreadis used, and we introduce a new metric, Community Entropy Ratio, to further extend the idea. Community Entropy Ratio seems to allow direct comparison across different graph topologies and shows encouraging potential for its ability to predict the eventual persistence of a Twitter hashtag.