Media Mediate Sentiments: Exploratory Analysis of Tweets Posted Before, During, and After the Great East Japan Earthquake

When the Great East Japan Earthquake occurred, Twitter was used as an infrastructure for sharing information carried by other media. In other words, Twitter is considered as a "meta medium." Earthquake-related tweets included information that was of questionable veracity, contained vicious rumors, and propagated matters of controversy that often gave rise to various discussions and arguments. In this research, the authors analyzed 89,351,242 tweets posted from December 11, 2010 to April 16, 2012. They then extracted 9,816,625 URLs and classified the top 100 domains of these URLs into 19 media categories. The emotional reactions of Twitter users were investigated by counting the terms conveying positive and negative emotions included in the body of tweets along with the media URLs. The authors' findings revealed differences in terms of the frequency with which terms expressing emotions were evoked and differences in the patterns of their surges, across the various media. The authors also considered the usage of various terms appearing in tweets concurrently with the terms expressing emotion.

[1]  Boris Egloff,et al.  The Emotional Timeline of September 11, 2001 , 2010, Psychological science.

[2]  J. Pennebaker,et al.  Linguistic Markers of Psychological Change Surrounding September 11, 2001 , 2004, Psychological science.

[3]  Yutaka Matsuo,et al.  Earthquake shakes Twitter users: real-time event detection by social sensors , 2010, WWW '10.

[4]  Taku Kudo,et al.  MeCab : Yet Another Part-of-Speech and Morphological Analyzer , 2005 .

[5]  B. Rimé Emotion Elicits the Social Sharing of Emotion: Theory and Empirical Review , 2009 .

[6]  P. Earle,et al.  OMG Earthquake! Can Twitter Improve Earthquake Response? , 2009 .

[7]  M Terasaki,et al.  [Construction of a multiple mood scale]. , 1992, Shinrigaku kenkyu : The Japanese journal of psychology.

[8]  R Monchi,et al.  [Development of the general affect scales]. , 2000, Shinrigaku kenkyu : The Japanese journal of psychology.

[9]  Scott A. Golder,et al.  Diurnal and Seasonal Mood Vary with Work, Sleep, and Daylength Across Diverse Cultures , 2011 .

[10]  Barbara Poblete,et al.  Twitter under crisis: can we trust what we RT? , 2010, SOMA '10.

[11]  Stefan Stieglitz,et al.  Emotions and Information Diffusion in Social Media—Sentiment of Microblogs and Sharing Behavior , 2013, J. Manag. Inf. Syst..

[12]  Yasuaki Sakamoto,et al.  The Impact of Posting URLs in Disaster-Related Tweets on Rumor Spreading Behavior , 2013, 2014 47th Hawaii International Conference on System Sciences.

[13]  Mike Thelwall,et al.  Sentiment in Twitter events , 2011, J. Assoc. Inf. Sci. Technol..

[14]  Chen Huang,et al.  Microblogging after a major disaster in China: a case study of the 2010 Yushu earthquake , 2011, CSCW.

[15]  Neil Savage,et al.  Twitter as medium and message , 2011, Commun. ACM.

[16]  Eiji Aramaki,et al.  Use trend analysis of twitter after the great east japan earthquake , 2012, CSCW.

[17]  Naohiro Matsumura,et al.  [Expression of negative emotional responses to the 2011 Great East Japan Earthquake: Analysis of big data from social media]. , 2015, Shinrigaku kenkyu : The Japanese journal of psychology.

[18]  Owen Rambow,et al.  Sentiment Analysis of Twitter Data , 2011 .

[19]  Atsushi Sato,et al.  Development of the Japanese version of Positive and Negative Affect Schedule (PANAS) scales , 2001 .

[20]  R Monchi,et al.  [The factor structure of relief feeling in situations: relief from tension and calmness and relaxation]. , 2000, Shinrigaku kenkyu : The Japanese journal of psychology.

[21]  Son Doan,et al.  An analysis of Twitter messages in the 2011 Tohoku Earthquake , 2011, eHealth.

[22]  Maite Taboada,et al.  Lexicon-Based Methods for Sentiment Analysis , 2011, CL.