Forecasting

In this project we developed and tested a first version of an early warning forecasting tool based on the FAST event database. The tool used sequence alignment algorithms to estimate conflict risk in monitored countries on a daily basis. A first version of the tool is available on the following website: http://fastserver.unibe.ch

Production of Event Data

swisspeace and JRC cooperated closely in this project to compare two different approaches to produce event data for conflict early warning purposes - event data generated by FAST's Local Information Networks and event data generated by automated parsing of Reuters news feed by Virtual Research Associates. In addition to an evaluation of these two different methods of producing event data, the project also described indicators useful in identifying countries vulnerable to conflict.

Information Sources of Event Data

In this project, we compared the informational content of the FAST event database with the English written part of the World Wide Web. The goal was to estimate the informational overlap between information gathered by Local Information Networks and information available on the WWW.

Network Analysis of Event Data

This project combined event data analysis with social network analysis. It demonstrated that the combination of these two approaches within a common framework has clear advantages. We have produced an actor-network analysis based on the FAST event database to identify the main actors involved in conflict, to track down the most important conflictive and cooperative ties between actors, and to compare the quality and intensity of interactions among actors. Results for Chechnya were published in the Journal of Conflict Resolution Vol. 50, No. 2. 159-175. 

Mapping of Event Data

swisspeace and JRC cooperated closely in this project to visualize data from the FAST event database on interactive topographic maps together with additional structural data. The goal of this project was to develop new techniques to communicate conflict relevant information and to gain a better understanding of spatial conflict dynamics. Such interactive maps, also called riskmaps, have already been produced for Kosovo, Armenia and Azerbaijan.