A Qualitative Examination of Topical Tweet and Retweet Practices

WSU CORE Repository

 

A Qualitative Examination of Topical Tweet and Retweet Practices

Show full item record

Preview: Thumbnail
Title: A Qualitative Examination of Topical Tweet and Retweet Practices
Author: Purohit, Hemant
Abstract:

Twitter's popularity in harnessing real-time traffic, enabling large-scale information diffusion and creating tangible effects on participating economies and societies is well known today. Just minutes after President Obama's address to Congress on healthcare, Twitter showed an avalanche of tweets about the outburst from Joe Wilson -and not inconsequentially, financial support began pouring into the campaign funds of both Joe Wilson and his opponent. Twitter's influence was also apparent following the terrorists attack in Mumbai, Haiti earthquake and in the civil reaction to the Iranian elections. Although it has been argued that, as with link-based blogging, re-tweeting (forwarding of a tweet) holds immense potential for viral marketing and content sharing, the mechanics of this practice is not well understood. We hypothesize that in addition to the social significance or timeliness of a tweet, there this is also a strong three-dimensional dynamic at play -the people involved (passionate advocate or an objective observer), the content being tweeted (fact-sharing or emotionally charged) and the connections between the people, all playa role in how a tweet spreads. Understanding these micro-level variables and their interactions will shed light on macro-level consequences e.g., political decisions or consumer behaviors. Our work contributes to the study of such retweet behavior on Twitter surrounding real-world events. We analyze over a million tweets pertaining to three events, present general tweet properties in such topical datasets and qualitatively analyze the properties of the retweet behavior surrounding the most tweeted/viral content pieces. Findings include a clear relationship between sparse/dense retweet patterns and the content and type of a tweet itself; suggesting the need to study content properties in link-based diffusion models.

This presentation occurred at the Wright State University Campus-Wide Celebration of Research, Scholarship and Creative Activities on April 16, 2010

Bookmark: http://hdl.handle.net/2374.WSU/4732
Date: April 2010

Files in this item

Files Size Format View
celebration_abstract10_purohit_h.pdf 87.20Kb application/pdf Thumbnail

This item appears in the following Collection(s)

Show full item record

Search CORE


Advanced Search

Browse

My Account

About

Links