| Full text | |
| Author(s): |
Vega-Oliveros, Didier A. A.
;
Koren, Ore
Total Authors: 2
|
| Document type: | Journal article |
| Source: | IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS; v. N/A, p. 11-pg., 2023-02-10. |
| Abstract | |
Civil war exhibits complex geospatial trends over time, which may be missed by models that rely on count-based operationalizations. Here, the spatial and temporal correlation values of monthly civil war events are transformed into their influence degree symbol, which measures geospatial concentration, spread, and intensity of civil war. We then measure variation in these degrees over time to identify relevant spatiotemporal civil war aspects. The network model is constructed using 0.5 degree grid locations as nodes, counting nearby and over time connections. We then extract the temporal community structure behind the data. We use ground-truth data to visualize how our measures correlate with observed patterns, thereby illustrating our method provides accurate depictions of geospatial civil war dynamics. We also evaluate the impact of several indicators highlighted by past research and our community-based spatiotemporal measures and comparing it to the preprocessed count indicator. Our findings indicate that the relationship between state capacity and climate stress show opposite correlations with civil war as those identified by studies that use count-based indicators. Counterintuitively, our results show that conflict intensifies and spreads in locations where the state is stronger and where climate conditions are improved. (AU) | |
| FAPESP's process: | 17/12646-3 - Déjà vu: feature-space-time coherence from heterogeneous data for media integrity analytics and interpretation of events |
| Grantee: | Anderson de Rezende Rocha |
| Support Opportunities: | Research Projects - Thematic Grants |
| FAPESP's process: | 18/24260-5 - Spatiotemporal Data Analytics based on Complex Networks |
| Grantee: | Didier Augusto Vega Oliveros |
| Support Opportunities: | Scholarships abroad - Research Internship - Post-doctor |
| FAPESP's process: | 19/26283-5 - Learning visual clues of the passage of time |
| Grantee: | Didier Augusto Vega Oliveros |
| Support Opportunities: | Scholarships in Brazil - Post-Doctoral |
| FAPESP's process: | 16/23698-1 - Dynamical Processes in Complex Network based on Machine Learning |
| Grantee: | Didier Augusto Vega Oliveros |
| Support Opportunities: | Scholarships in Brazil - Post-Doctoral |