Sign In

Communications of the ACM

ACM TechNews

Twitter Data Crunching: The New Crystal Ball

View as: Print Mobile App Share:
Twitter logo


Big data analysis of Twitter activity can help predict the outcome of election-based processes such as TV talent shows, according to new research.

Northeastern University's Fabio Ciulla and colleagues were able to anticipate the elimination of contestants on shows based on public voting, such as American Idol, using tweets. Twitter offers a quantitative indicator that can act as a proxy for what is occurring in the world in real time. The team was able to collect very detailed data on social behavior on a massive scale.

The approach provides a simplified version of analyzing complex societal phenomena such as political elections. Ciulla and colleagues note that some of the tweets included geolocalization information, which made it possible to internationally map the fan base of each contestant. Moreover, Twitter offers a way to study social phenomena in a completely unobtrusive way.

From AlphaGalileo
View Full Article

Abstracts Copyright © 2012 Information Inc., Bethesda, Maryland, USA 


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account