IDEAS home Printed from https://ideas.repec.org/a/taf/ginixx/v48y2022i4p800-822.html
   My bibliography  Save this article

Conflict forecasting with event data and spatio-temporal graph convolutional networks

Author

Listed:
  • Patrick T. Brandt
  • Vito D’Orazio
  • Latifur Khan
  • Yi-Fan Li
  • Javier Osorio
  • Marcus Sianan

Abstract

This paper explores three different model components to improve predictive performance over the ViEWS benchmark: a class of neural networks that account for spatial and temporal dependencies; the use of CAMEO-coded event data; and the continuous rank probability score (CRPS), which is a proper scoring metric. We forecast changes in state based violence across Africa at the grid-month level. The results show that spatio-temporal graph convolutional neural network models offer consistent improvements over the benchmark. The CAMEO-coded event data sometimes improve performance, but sometimes decrease performance. Finally, the choice of performance metric, whether it be the mean squared error or a proper metric such as the CRPS, has an impact on model selection. Each of these components–algorithms, measures, and metrics–can improve our forecasts and understanding of violence.En este artículo se exploran tres componentes diferentes del modelo para mejorar el rendimiento predictivo con respecto a la referencia ViEWS: una clase de redes neuronales que tienen en cuenta las dependencias espaciales y temporales, el uso de datos de eventos codificados por CAMEO, y la puntuación de probabilidad de rango continuo (CRPS), que es una métrica de puntuación adecuada. Predecimos los cambios en la violencia estatal en toda África a nivel mensual. Los resultados muestran que los modelos de redes neuronales convolucionales de gráficos espacio-temporales ofrecen mejoras consistentes sobre el punto de referencia. Los datos de eventos codificados por CAMEO a veces mejoran el rendimiento, pero otras veces lo empeoran. Por último, la elección de la métrica de rendimiento, ya sea el error cuadrático medio o una métrica propia como la CRPS, influye en la selección del modelo. Cada uno de estos componentes (algoritmos, medidas y métricas) puede mejorar nuestras previsiones y nuestra comprensión de la violencia.Cet article explore trois composantes de modèles différentes pour améliorer les performances prédictives par rapport à la référence de ViEWS (Violence early-warning system, système d’alerte précoce sur la violence) : une classe de réseaux de neurones qui prennent en compte les dépendances spatiales et temporelles ; l’utilisation de données d’événements codées par CAMEO (Conflict and Mediation Events Observations, Observation des événements de médiation et de conflit) ; et le CRPS (Continuous Rank Probability Score, Score de probabilité de catégories ordonnées de variables continues), qui est une métrique de score propre. Nous effectuons des prédictions des évolutions de la violence étatique en Afrique au niveau grille/mois. Les résultats montrent que les modèles à réseaux convolutifs de neurones graphiques spatiotemporels offrent des améliorations constantes par rapport à la référence. Les données d’événements codées par CAMEO améliorent parfois les performances mais peuvent aussi parfois les réduire. Enfin, le choix de la métrique de performances, qu’il s’agisse de l’erreur quadratique moyenne ou d’une métrique de score propre telle que le CRPS, a un impact sur la sélection du modèle. Chacune de ces composantes - algorithmes, mesures et métriques - peut améliorer nos prévisions et notre compréhension de la violence.

Suggested Citation

  • Patrick T. Brandt & Vito D’Orazio & Latifur Khan & Yi-Fan Li & Javier Osorio & Marcus Sianan, 2022. "Conflict forecasting with event data and spatio-temporal graph convolutional networks," International Interactions, Taylor & Francis Journals, vol. 48(4), pages 800-822, July.
  • Handle: RePEc:taf:ginixx:v:48:y:2022:i:4:p:800-822
    DOI: 10.1080/03050629.2022.2036987
    as

    Download full text from publisher

    File URL: http://hdl.handle.net/10.1080/03050629.2022.2036987
    Download Restriction: Access to full text is restricted to subscribers.

    File URL: https://libkey.io/10.1080/03050629.2022.2036987?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    As the access to this document is restricted, you may want to search for a different version of it.

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Racek, Daniel & Thurner, Paul & Kauermann, Goeran, 2024. "Integrating Spatio-temporal Diffusion into Statistical Forecasting Models of Armed Conflict via Non-parametric Smoothing," OSF Preprints q59dr, Center for Open Science.

    More about this item

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:taf:ginixx:v:48:y:2022:i:4:p:800-822. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    We have no bibliographic references for this item. You can help adding them by using this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Chris Longhurst (email available below). General contact details of provider: http://www.tandfonline.com/GINI20 .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.