Measuring the Public Mood State from Large-Scale Microblogging

Johan Bollen 

Wednesdays@NICO Seminar, Noon, January 19 2011, Chambers Hall, Lower Level

Prof. Johan Bollen, Indiana University School of Informatics and Computing

Abstract

We perform a sentiment analysis of all public tweets broadcast by Twitter users. For every day in the timeline, we extract six dimensions of mood (tension, depression, anger, vigor, fatigue, confusion) using an extended version of the Profile of Mood States (POMS), a well-established psychometric instrument. We compare our results to fluctuations recorded by the stock market and major socio-economic events, such as the U.S. Presidential Election of November 4, 2008 and Thanksgiving Day. We find that socio-economic events have a significant, immediate and highly specific effect on the various dimensions of public mood. Long-term changes in the public mood state reflect the cumulative effect of various underlying socio-economic indicators.

Biography

Johan Bollen is associate professor at the Indiana University School of Informatics and Computing (2009-present). He was formerly a staff scientist at the Los Alamos National Laboratory (2005-2009). His research takes place at the intersection between informatics, computer science, network science, social science, and informetrics. His present interests include the develop-ment of quantitative tools for the assessment of scholarly impact from usage data (MESUR project) and the study of social phenomena such as public mood states and social contagion from large-scale online networking environments.  More information can be found at http://informatics.indiana.edu/jbollen/Home.html