MAPPING WIKIPEDIA: DATA MINING, GEOCODING AND SPATIAL ANALYSIS OF ARCHIVED EDIT HISTORY (2001-2009)

Anyone in the world with a computer and Internet-access can type www.wikipedia.com and be welcomed to “the free encyclopedia that anyone can edit.” With a simple click anyone can “edit this page” and participate in the Wikipedia project. While this encourages broad participation, it also invites controversy among editors with conflicting interpretations of events or topics. Although site rules hold that users must develop a neutral point-of-view through consensus decision-making, the process is fundamentally biased towards the most prevalent political, economic, and cultural perspectives on the site. Using IP addresses to map the geography of Wikipedia editors, we can explore political, economic, and cultural divides on Wikipedia and develop strategies to bridge these divides.

This research focuses on an historical log of edits to Wikipedia. This log links every edit to a specific user, physical location, contribution, date and time. By studying this log it is possible to learn about the geography and politics of Wikipedia editors. Because this log contains hundreds of millions of entries, this research must make use of computational methods such as data mining in addition to traditional social scientific methods. Data mining makes possible the analysis of this huge volume of data.

My findings contrast with previous studies that focus on Wikipedia as a site of collaboration and consensus. This study illuminates a fundamental bias on Wikipedia that is supported by a hegemonic consensus, and also illustrates the geography of this bias.