This repository contains all code files for reproducing and/or replicating my presentation at the 64th conference of the International Linguistic Association.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Wikipedia (Main Page, n.d.) has grown to become a go-to source for all kinds of information available in multiple languages. It is often and erroneously seen as an unbiased source since it is created by its users as an open-source project and is democratically edited. While a wealth of information can indeed be seen as unbiased, bias is frequently present when articles on controversial topics are published. Wikipedia allows any user to edit a given article by clicking on the "edit" button, which takes her/him to a new page where the text can be altered. Thus, articles on Wikipedia typically contain several revisions, back-and-forth, until a consensus is reached. When more contentious issues are featured, however, a consensus may never be reached, and the competing factions continue their battle for their truth to become the established one. This study analyzes Wikipedia articles related to the Catalonian Independence Process (El Procés), a movement to separate from Spain, ultimately bringing about a contested referendum on independence in 2017. It is shown that this controversy has resulted in hundreds of competing edits on the related pages. Using multidimensional sentiment analysis (Jockers, 2015), we can further observe how sentiments have changed over time and are affected by discrete historical events. Quantitative and qualitative analyses of the contributions also show that the competing factions have deployed artificial intelligence in their quest for an upper hand in defense of their position. A comparison of articles across languages –versions are available in Spanish, Catalan, Portuguese, Basque, Galician and Esperanto– show that users of minority languages tend to be more sympathetic to the independence movement. This is perhaps not surprising given the history of Spain and the current context. We also see that, while the positions may converge/diverge over time, the truth mof the matter is most often divergent, La Verdad is not La Veritat.
Pull requests are welcome. Also, please feel free to contact me at: dietrichson@gmail.com