Working Paper: CEPR ID: DP13748
Authors: Hannes Felix Mueller; Christopher Rauh
Abstract: There is a growing interest in better conflict prevention and this provides a strong motivation for better conflict forecasting. A key problem of conflict forecasting for prevention is that predicting the start of conflict in previously peaceful countries is extremely hard. To make progress in this hard problem this project exploits both supervised and unsupervised machine learning. Specifically, the latent Dirichlet allocation (LDA) model is used for feature extraction from 3.8 million newspaper articles and these features are then used in a random forest model to predict conflict. We find that forecasting hard cases is possible and benefits from supervised learning despite the small sample size. Several topics are negatively associated with the outbreak of conflict and these gain importance when predicting hard onsets. The trees in the random forest use the topics in lower nodes where they are evaluated conditionally on conflict history, which allows the random forest to adapt to the hard problem and provides useful forecasts for prevention.
Keywords: armed conflict; forecasting; newspaper text; machine learning; topic models; random forest
JEL Codes: No JEL codes provided
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| justice topics (K40) | conflict risk (D74) |
| diplomacy topics (F51) | conflict risk (D74) |
| economics topics (A10) | conflict risk (D74) |
| conflict history (D74) | future conflict predictions (D74) |
| effective forecasting (C53) | proactive measures against potential conflicts (D74) |