Understanding the Initiation, Escalation, and Termination of Interstate Confrontations

University Of Georgia Research Foundation Inc, Athens GA

Investigators

Abstract

This project develops theory about confrontation initiation, escalation, and termination. It generates new data to assess predictions from this theory. This involves the use of state-of-the-art advancements in artificial intelligence (AI) to train a model that uses a large text corpus to identify events and to describe a range of characteristics for each event. The project will shed new light on how confrontations begin, why some minor confrontations become major conflicts, and what can be done to end conflicts. The data allows real-time descriptions of characteristics as confrontations unfold, which will be use for analysis and pedagogy. This project contributes new theory and data to the study of militarized interstate confrontations. It introduces new theory about the conditions under which confrontations emerge, why they increase in intensity, and how they end. To create a new dataset on confrontations, it first develops and cleans a large text corpora. Each element of the corpora with relevant text is attributed a feature or feature of the dataset. This labelled dataset is then used to fine-tune a BERT classifier, which allows the artificial intelligence model to identify and code subsequent entries automatically. The dataset allows testing of propositions derived from the theory developed in the project, can also be used by other scholars to investigate additional hypotheses, and can be updated in real-time for analysis. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.