方法对比

	Machine Learning Conflict Prediction	Dyadic Conflict Analysis
领域	International Relations	International Relations
方法族	Machine learning	Process / pipeline
起源年份	2016	1992
提出者	Predictive conflict research (e.g., Muchlinski, Siroky, He & Kocher)	Stuart A. Bremer (and the Correlates of War dyadic tradition)
类型	Supervised machine-learning prediction of conflict	Observational research design for interstate conflict
开创性文献	Muchlinski, D., Siroky, D., He, J., & Kocher, M. (2016). Comparing random forest with logistic regression for predicting class-imbalanced civil war onset data. Political Analysis, 24(1), 87–103.	Bremer, S. A. (1992). Dangerous dyads: Conditions affecting the likelihood of interstate war, 1816–1965. Journal of Conflict Resolution, 36(2), 309–341.
别名	ML Conflict Prediction, Random Forest Civil War Prediction, Algorithmic Conflict Prediction, Supervised Learning for Conflict Onset	Dyad-Year Analysis, Dyadic Design in Conflict Studies, Dangerous Dyads Analysis, Pairwise Interstate Conflict Analysis
相关	3	3
摘要	Machine learning conflict prediction uses flexible supervised algorithms — random forests, gradient boosting, neural networks, regularized regression — to forecast the onset of armed conflict from large sets of features, prioritizing out-of-sample predictive accuracy over coefficient interpretation. Muchlinski, Siroky, He, and Kocher (2016) showed that random forests substantially outperform logistic regression at predicting class-imbalanced civil-war onset, catalyzing a shift in conflict research toward algorithmic prediction, rigorous out-of-sample validation, and the recognition that explanation and prediction are distinct goals.	Dyadic conflict analysis is the dominant research design in quantitative conflict studies: it treats the pair of states (the dyad), observed year by year, as the unit of analysis and models the probability that a pair experiences militarized conflict as a function of their joint and individual attributes. Stuart Bremer's 'Dangerous Dyads' (1992) is the canonical statement, identifying which conditions — contiguity, the absence of alliance, power parity, the absence of joint democracy, and others — make a pair of states war-prone. The design aligns conflict data with the relational theories that dominate the field.