Swarm Creativity Blog

In earlier work we have constructed people maps based on their linking structure in different-language Wikipedias. Articles about people in Wikipedia include many social cues about a particular person. By using this information, we create a social network of people listed in the Wikipedia article. We only look at links from one people page to another people page. The links between these pages convey a rich social context, such as friendship, animosity, communication, or having participated in the same events, for example. The meaning of a link includes relationship information about a person, for instance, with whom they have worked as well as basic demographic information such as when and where the person was born. We have compared the English/ German/ French /Japanese/ Chinese/ Spanish/ Korean people maps, with substantial differences on who matters in what culture. In the meantime, we have been lucky that Sar Haidar , a native Arabic speaker, agreed to repeat the analysis for the A...

Yesterday Nielsenwire published a small study Buzz and the Ballot Box: What is Social Media’s Relationship with Politics? where they compared Web Buzz with election success of politicians. Nielsen analyzed 4 local elections: - Senate – California: Barbara Boxer vs Carly Fiorina - Senate – Florida: Marco Rubio vs Kendrick Meek vs Charlie Christ - Governor – Ohio: John Kasich vs Ted Strickland - Governor – Maryland: Martin O’Malley vs Bob Erlich by comparing the posts on blogs, blog comments, online news sites, video and image sites, message boards/groups, and public posts on Twitter with election success. They found that for 3 out of the 4 Web Buzz predicted outcome correctly. It's nice to get confirmation of what I had found for the last six years from other sources! Thanks to Sar Haidar for pointing out the article.