Comparer des méthodes

	Civil War Onset Analysis	Machine Learning Conflict Prediction
Domaine	International Relations	International Relations
Famille	Process / pipeline	Machine learning
Année d'origine	2003	2016
Auteur d'origine	Civil-war research program (e.g., James Fearon & David Laitin; Collier & Hoeffler)	Predictive conflict research (e.g., Muchlinski, Siroky, He & Kocher)
Type	Observational country-year analysis of civil-war onset	Supervised machine-learning prediction of conflict
Source fondatrice	Fearon, J. D., & Laitin, D. D. (2003). Ethnicity, insurgency, and civil war. American Political Science Review, 97(1), 75–90.	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.
Alias	Civil Conflict Onset Analysis, Greed vs. Grievance Analysis, Insurgency Onset Analysis, Determinants of Civil War	ML Conflict Prediction, Random Forest Civil War Prediction, Algorithmic Conflict Prediction, Supervised Learning for Conflict Onset
Apparentées	3	3
Résumé	Civil war onset analysis is the observational study of why internal armed conflict begins in some countries and years but not others. Organized as country-year data with a binary onset outcome, it models the probability of onset against structural, economic, and political conditions. Fearon and Laitin's (2003) influential analysis argued that civil war is best understood as insurgency, and that the conditions favoring insurgency — weak states, poverty, rough terrain, large populations — predict onset far better than ethnic or religious diversity, reframing the long 'greed versus grievance' debate.	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.