Visualizing and monitoring emergent border issues in Central Asia

Twenty years after their independence, the five countries of Central Asia face a number of issues, which can be observed at their borders, inherited from the Soviet era. The borders demarcating one from another have been enforced as a result of the formation of new national identities. In contrast, more (official or smuggled) exchanges are taking place across the borders of the former USSR. Internet-based media specialized in Central Asia regularly report on events, conflicts and exchanges occurring in the area. They provide a considerable body of information in the form of press releases, which can help the researcher to understand how these events are triggered or interrelated. However, the amount of text rapidly becomes excessive for an individual researcher. Our approach is to assemble these reports in a text corpus and process them with text mining techniques, allowing both information extraction and text categorization. Our text corpus is built with reports (covering a period of 11 years) issued by the Institute for War and Peace Reporting (http://iwpr.net), which produces homogeneous and regular information based on reliable and neutral journalism sourcing. With the release dates of the reports it is possible to work on time series and if the corpus is enhanced with the ability to handle easy updates it is possible to monitor current events.