К ранжированию значимости факторов дестабилизации в странах Азии и Африки методами машинного обучения

Восток. Афро-азиатские общества: история и современность. 2026. № 3. С. 117–130.

Korotayev A., Chernomorchenko I., Медведев И. А.

This study employs machine learning methods to rank factors contributing to large-scale armed and unarmed destabilization across Asian and African countries. Analysis reveals that African nations demonstrate greater vulnerability to armed destabilization (up to full-scale civil wars), whereas Asian countries are more prone to less violent unarmed forms (mass antigovernment demonstrations, riots, general strikes and so on). Key predictors of armed destabilization – beyond the dominant factor of prior armed civil conflict – include population size, US foreign aid inflows, GDP per capita decline, and resource extraction (oil, gold). Crucially, low state effectiveness index emerges as a significant predictor exclusively for armed destabilization in Asia, while low cereal yields prove uniquely significant in Africa. For nonviolent destabilization, primary factors comprise low GDP per capita growth, poor cereal yields, expanded coverage of population with modern formal education, corruption, and political variables (incumbent duration, intermediate values of electoral democracy index). Notably, in Asia, the probability of unarmed protest destabilization is increased by low ruling party legitimacy, party system fragmentation, judicial corruption, and sharp export contractions. In Africa, risks of unarmed mass protest are increased by low military expenditure per capita, unusually high cereal yields, abrupt import reductions, and systemic corruption. Furthermore, expanded educational coverage reduces armed destabilization risks but amplifies less violent unarmed destabilization probability in both regions. Political regime analysis indicates that young democracies in Africa exhibit heightened vulnerability to armed destabilization, whereas hybrid regimes in Asia are more susceptible to nonviolent forms. Corruption turns out to predict nonviolent rather than armed destabilization.

Keywords: машинное обучение, social factors, социальные факторы, дестабилизация, destabilization, страны Азии, страны Африки, экономические факторы, political factors, политические факторы, African countries, economic factors, machine learning, Asian Countries

Publication based on the results of: State Resilience and National Security of Russia in the Context of the World Order Transformation (2025)