Computational approaches to conflict research from modeling and data to computational diplomacy

This paper offers an account of our own efforts to draw on computational methods to study conflict processes at the macro level. During a first phase, we relied on agent-based modeling in order to capture the complexity of system-level processes. This research yielded a number of publications, but less by way of influence on substantive research and policy making. Therefore, we decided to shift our main focus away from agent-based modeling to spatial computation, which allow for a more direct empirical validation of our results. This second phase of research includes the collection and integration of large amounts of spatiotemporally structured data, which we analyze with more conventional econometric tools. To advance the field of computational diplomacy, we recommend that future search combines agent-based modeling with rigorous empirical validation through the utilization of spatial computation.