Tracking Horizontal Inequalities Across Dimensions to Forecast and Understand Instability (TrIAD)

点击翻译按钮获取中文摘要

Inequality within and across societies is at the forefront of academic and policy discussions related to global-scale challenges from the rise of violent extremism, the escalation and spread of civil conflict, and potential solutions to environmental shocks. New conceptual research suggests that group-based grievances about access to rights and resources, known as horizonal inequalities (HIs), as compared to individual-to-individual vertical inequalities, are crucial mechanisms that explain ethnic conflict and instability. However, the complexity of measuring horizontal inequalities across different and overlapping groups, rights and resources, and societal contexts, has impeded scientific understanding and policy-makers’ ability to intervene and mitigate the potentially destabilizing consequences of HIs. TrIAD (Tracking Horizontal Inequalities Across Dimensions to Forecast and Understand Instability) overcomes previous hurdles by linking scholars from the social, information, and computer sciences, as well as drawing on expertise related to political, security, and historical dimensions. Furthermore, the project provides a fertile environment for students with diverse backgrounds to gain high-quality research experience and develop highly sought-after skills in data science, in general, and computational social science in particular. The knowledge and technologies created from this project lay the foundation for the sustained development of skills for the emerging data-driven society.TrIAD (Tracking Horizontal Inequalities Across Dimensions to Forecast and Understand Instability) uniquely connects work on human rights measurement to the concept of horizonal inequalities (HIs). TrIAD particularly harnesses the sparse and streaming large-scale data on human rights, using advances in information-science methods, Bayesian models, and high-performance computing, to understand how and where horizonal inequalities lead to escalating instability and discover new means of counteracting this violence. Using PULSAR, a human rights text parser, TrIAD is able to extract structured information from the streams of raw textual information authored and published by hundreds of human rights non-governmental organization in press releases and formal reports daily. From these human rights documents, TrIAD accurately identifies the perpetuators and specific victims, as well as codes the human right/resource that is being violated. TrIAD also includes a modeling and analysis framework that infers the latent structure of grievances and how they project into violence across different contexts. The completed TrIAD provides researchers and policy-makers with highly disaggregated data on systematically unequal access to over 100 resources and rights, across different societal groupings, including salient racial, linguistic, religious, regional, clan, and ideological partitions that vary over time and space.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.