International Interactions最新文献

{"title":"Predicting (de-)escalation of sub-national violence using gradient boosting: Does it work?","authors":"Jonas Vestby, Jürgen Brandsch, V. B. Larsen, Peder Landsverk, A. Tollefsen","doi":"10.1080/03050629.2022.2021198","DOIUrl":"https://doi.org/10.1080/03050629.2022.2021198","url":null,"abstract":"Abstract This article presents a prediction model of (de-)escalation of sub-national violence using gradient boosting. The prediction model builds on updated data from the PRIO-GRID data aggregator, contributing to the ViEWS prediction competition by predicting changes in violence levels, operationalized using monthly fatalities at the 0.5 × 0.5-degree grid (pgm) level. Our model's predictive performance in terms of mean square error (MSE) is marginally worse than the ViEWS baseline model and inferior to most other submissions, including our own supervised random forest model. However, while we knew that the model was comparatively worse than our random forest model in terms of MSE, we propose the gradient boosting model because it performed better where it matters—in predicting when (de-)escalation happens. This choice means that we question the usefulness of using MSE for evaluating model performance and instead propose alternative performance measurements that are needed to understand the usefulness of predictive models. We argue that future endeavors using this outcome should measure their performance using the Concordance Correlation, which takes both the trueness and the precision elements of accuracy into account, and, unlike MSE, seems to be robust to the issues caused by zero inflation. Este artículo presenta un modelo de predicción de la desescalada de la violencia subnacional mediante el uso de la potenciación del gradiente. El modelo de predicción se basa en los datos actualizados que provienen del agregador de datos de PRIO-GRID, contribuye al concurso de predicciones de ViEWS al predecir cambios en los niveles de violencia y es operacionalizado utilizando las muertes mensuales a nivel de cuadrícula de 0.5 × 0.5 grados (pgm). El rendimiento predictivo de nuestro modelo desde el punto de vista del error cuadrático medio (mean square error, MSE) es ligeramente peor que el modelo de referencia del sistema de alerta temprana sobre la violencia (Violence Early Warning System, ViEWS) e inferior en relación con la mayoría de las otras presentaciones, incluido nuestro modelo de bosque aleatorio y supervisado. No obstante, si bien sabíamos que el modelo era comparativamente peor que nuestro modelo de bosque aleatorio en relación con el MSE, proponemos el modelo de potenciación del gradiente porque funcionó mejor en el aspecto que importa: predecir cuándo ocurre la desescalada. Esta elección significa que cuestionamos la utilidad del uso del MSE para evaluar el rendimiento del modelo y, en cambio, proponemos mediciones de rendimiento alternativas que son necesarias para comprender la utilidad de los modelos predictivos. Sostenemos que, en los futuros proyectos en los que se utilice este resultado, se debería medir el rendimiento mediante la correlación de concordancia, la cual tiene en cuenta tanto los elementos de veracidad como los de precisión de la exactitud y, a diferencia del MSE, parece ser resistente a los problemas generados por la in","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"841 - 859"},"PeriodicalIF":1.3,"publicationDate":"2022-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42804543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"United they stand: Findings from an escalation prediction competition","authors":"Paola Vesco, Håvard Hegre, Michael Colaresi, R. Jansen, Adeline Lo, Gregor Reisch, Nils B. Weidmann","doi":"10.1080/03050629.2022.2029856","DOIUrl":"https://doi.org/10.1080/03050629.2022.2029856","url":null,"abstract":"Abstract This article presents results and lessons learned from a prediction competition organized by ViEWS to improve collective scientific knowledge on forecasting (de-)escalation in Africa. The competition call asked participants to forecast changes in state-based violence for the true future (October 2020–March 2021) as well as for a held-out test partition. An external scoring committee, independent from both the organizers and participants, was formed to evaluate the models based on both qualitative and quantitative criteria, including performance, novelty, uniqueness, and replicability. All models contributed to advance the research frontier by providing novel methodological or theoretical insight, including new data, or adopting innovative model specifications. While we discuss several facets of the competition that could be improved moving forward, the collection passes an important test. When we build a simple ensemble prediction model—which draws on the unique insights of each contribution to differing degrees—we can measure an improvement in the prediction from the group, over and above what the average individual model can achieve. This wisdom of the crowd effect suggests that future competitions that build on both the successes and failures of ours, can contribute to scientific knowledge by incentivizing diverse contributions as well as focusing a group’s attention on a common problem. Este artículo presenta los resultados y las enseñanzas extraídas en el marco de un certamen de predicción organizado por los responsables del proyecto Sistema de Alerta Temprana de Violencia (Violence Early-Warning System, ViEWS) con el propósito de mejorar los conocimientos científicos colectivos sobre la previsión de la (des)escalada en el continente africano. En el certamen se pidió a los participantes que desarrollaran una previsión con respecto a los cambios en la violencia estatal para el futuro real (de octubre de 2020 a marzo de 2021), así como para una muestra de prueba que se mantendría. Se formó un comité de calificación externo, independiente tanto de los organizadores como de los participantes, para evaluar los modelos en función de criterios cualitativos y cuantitativos, como el rendimiento, la novedad, la singularidad y la replicabilidad. Todos los modelos contribuyeron a avanzar en la frontera de la investigación mediante el aporte de nuevos conocimientos metodológicos o teóricos, la inclusión de nuevos datos o la adopción de especificaciones innovadoras del modelo. Aunque se debarió sobre varios aspectos del certamen que podrían mejorarse de cara al futuro, lo que se recopiló pasó una prueba importante. Cuando se construye un simple modelo de predicción de conjunto, que se basa en los conocimientos únicos de cada contribución en diferentes grados, se puede medir una mejora en la predicción del grupo, por encima de lo que el modelo individual promedio puede lograr. Este efecto de la sabiduría de la multitud sugiere que los futuros cert","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"860 - 896"},"PeriodicalIF":1.3,"publicationDate":"2022-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43837463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Why do states contribute to the global refugee governance? Fiscal burden-sharing in the post-2011 Syrian refugee crisis","authors":"Hirotaka Fujibayashi","doi":"10.1080/03050629.2022.2040496","DOIUrl":"https://doi.org/10.1080/03050629.2022.2040496","url":null,"abstract":"Abstract Why are some states motivated to contribute financially to international efforts to protect refugees and assist host countries? Despite general agreement on the need for burden-sharing in global refugee governance, research up to now has not explained the underlying motivations behind the provision of financial assistance tied to the international protection of refugees. In addressing this gap, this article offers two competing perspectives of the potential impact of refugee migration on the decisions by individual states concerning whether and how much they contribute to a given refugee crisis. The article further hypothesizes that the connection between refugee migration and states’ financial contribution depends on the geographic context. The proposed hypotheses are tested using the cross-country panel data on humanitarian assistance to the post-2011 Syrian refugee crisis, and the test confirms that states receiving a more significant number of refugees have a greater incentive to offer a financial contribution. However, this explanation only holds for contributing states remote from Syria. Conversely, states in the geographical proximity of Syria likely have fewer interests to take on a financial burden to support Syrian refugees staying outside of their territories. These findings provide several important insights into the broader policy of refugee governance and into academic debates on the sharing of financial burdens to protect refugees. ¿Por qué algunos estados tienen la motivación de aportar dinero a los esfuerzos internacionales para proteger a los refugiados y ayudar a los países anfitriones? A pesar del acuerdo general sobre la necesidad de compartir la carga en la gestión global de los refugiados, la investigación realizada hasta ahora no ha explicado las motivaciones subyacentes a la provisión de ayuda financiera relacionada con la protección internacional de los refugiados. A fin de abordar esta brecha, en el presente artículo, se ofrecen dos perspectivas contrapuestas sobre el impacto potencial de la migración de los refugiados en las decisiones de los estados individuales respecto a si contribuyen o no, y en qué medida, a una determinada crisis de refugiados. En el artículo, también se plantea la hipótesis de que la conexión entre la migración de refugiados y la contribución financiera de los estados depende del contexto geográfico. Las hipótesis propuestas se ponen a prueba utilizando los datos del panel de países sobre la asistencia humanitaria a la crisis de refugiados sirios después de 2011, y la prueba confirma que los estados que reciben una cantidad más considerable de refugiados tienen un mayor incentivo para ofrecer una contribución financiera. Sin embargo, esta explicación solo es válida para los estados contribuyentes alejados de Siria. Por el contrario, los estados situados en la proximidad geográfica de Siria probablemente estén menos interesados en asumir una carga financiera para ayudar a los refugiados ","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"345 - 373"},"PeriodicalIF":1.3,"publicationDate":"2022-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43253817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Labor rights in comparative perspective: The WorkR dataset","authors":"Colin M. Barry, David Cingranelli, K. Clay","doi":"10.1080/03050629.2022.2040495","DOIUrl":"https://doi.org/10.1080/03050629.2022.2040495","url":null,"abstract":"Abstract We present new data on labor laws and practices for all countries of the world, recorded annually from 1994 to 2010. The dataset covers seven different labor standards. These are freedom of association, collective bargaining, minimum wage, limitations on working hours, protection from unsafe and unhealthy working conditions, protection from forced and compulsory labor, and protection of children and young people. This list reflects an array of internationally recognized labor rights and closely corresponds with the standards identified in the ILO’s 1998 Declaration on Fundamental Principles and Rights at Work. A simple analysis demonstrates the utility of the dataset and showcases some of the ways it might be used as a powerful tool in the scientific study of labor rights specifically, and human rights more generally. Presentamos nuevos datos sobre leyes y prácticas laborales para todos los países del mundo. Estos datos se registraron anualmente desde 1994 hasta 2010. El conjunto de datos abarca siete normas laborales diferentes. Se trata de la libertad de asociación, la negociación colectiva, el salario mínimo, la limitación de la jornada laboral, la protección contra las condiciones de trabajo inseguras e insalubres, la protección contra el trabajo forzado y obligatorio, y la protección de los niños y los jóvenes. Esta lista incluye una serie de derechos laborales reconocidos internacionalmente y se corresponde estrechamente con las normas identificadas en la Declaración sobre los Principios y Derechos Fundamentales en el Trabajo de la OIT de 1998. Mediante un sencillo análisis, se demuestra la utilidad del conjunto de datos y se muestran algunas de las formas en que podría utilizarse como una herramienta eficaz en el estudio científico de los derechos laborales en particular, y de los derechos humanos en general. Nous présentons de nouvelles données sur les lois et pratiques du travail pour tous les pays du monde qui ont été enregistrées annuellement de 1994 à 2010. Le jeu de données couvre sept normes du travail différentes. Elles comprennent la liberté syndicale, les négociations collectives, le salaire minimum, la limitation du temps de travail, la protection contre les conditions de travail dangereuses et insalubres, la protection contre le travail forcé et obligatoire et la protection des enfants et des jeunes. Cette liste reflète toute une série de droits du travail internationalement reconnus et correspond étroitement aux normes identifiées dans la Déclaration de l’OIT relative aux principes et droits fondamentaux au travail de 1998. Une analyse simple démontre l’utilité de ce jeu de données et présente certaines des façons dont il pourrait être utilisé en tant qu’outil puissant dans l’étude scientifique des droits du travail en particulier, et des droits de l’homme de manière plus générale.","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"327 - 344"},"PeriodicalIF":1.3,"publicationDate":"2022-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43307509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Initiator conditions and the diffusion of digital trade-related provisions in PTAs","authors":"Manfred Elsig, S. Klotz","doi":"10.1080/03050629.2022.2004137","DOIUrl":"https://doi.org/10.1080/03050629.2022.2004137","url":null,"abstract":"ABSTRACT Digital trade has become an important driver of global commerce and accounts for an increasing share of many countries’ economies. While progress in digital trade-related discussions at the World Trade Organization has been limited until fairly recently, the topic has gradually been gaining importance in preferential trade agreements (PTAs) since the early 2000s. As we also observe that digital trade governance has become increasingly politicized, we know little about these provisions’ origins and diffusion in PTAs. This research note discusses novel data and analyzes 91 digital trade-related provisions and 347 trade agreements signed between 2000 and 2019. In this note, we focus primarily on the initiator conditions and how these might lead to differences in diffusion patterns. We find that almost half of digital trade-related provisions were initially introduced by PTAs in which the United States was a signatory. Using negative binomial regressions, we find no evidence, however, that these provisions diffuse relatively more often than provisions first introduced by other countries. Our analysis shows that the diffusion of digital trade-related provisions is influenced by original trade interests and the existence of domestic digital policies at the initiator stage. Interestingly, we find that the initial degree of legalization of the provisions themselves matters for a more substantial diffusion, which contradicts the established view that soft law provisions are the preferred approach for new trade topics. This research note highlights the need to factor in the extent to which new international law obligations are adopted through the treaty networks as a result of initiator conditions addressing a certain blind spot in the diffusion literature. By focusing on the initiating states, we also speak to the literature on how international agreements serve to diffuse leading states’ preferred policy options.","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"292 - 308"},"PeriodicalIF":1.3,"publicationDate":"2022-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44186882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Race to the bottom: Spatial aggregation and event data","authors":"S. Cook, Nils B. Weidmann","doi":"10.1080/03050629.2022.2025365","DOIUrl":"https://doi.org/10.1080/03050629.2022.2025365","url":null,"abstract":"Abstract Researchers now have greater access to granular georeferenced (i.e., spatial) data on social and political phenomena than ever before. Such data have seen wide use, as they offer the potential for researchers to analyze local phenomena, test mechanisms, and better understand micro-level behavior. With these political event data, it has become increasingly common for researchers to select the smallest spatial scale permitted by the data. We argue that this practice requires greater scrutiny, as smaller spatial or temporal scales do not necessarily improve the quality of inferences. While highly disaggregated data reduce some threats to inference (e.g., aggregation bias), they increase the risk of others (e.g., outcome misclassification). Therefore, we argue that researchers should adopt a more principled approach when selecting the spatial scale for their analysis. To help inform this choice, we characterize the aggregation problem for spatial data, discuss the consequences of too much (or too little) aggregation, and provide some guidance for applied researchers. We demonstrate these issues using both simulated experiments and an analysis of spatial patterns of violence in Afghanistan. Los investigadores tienen ahora un acceso como nunca antes a datos georreferenciados granulares (es decir, espaciales) sobre fenómenos sociales y políticos. Estos datos se han utilizado ampliamente, ya que ofrecen a los investigadores la posibilidad de analizar fenómenos locales, probar mecanismos y comprender mejor el comportamiento a nivel micro. Con estos datos sobre acontecimientos políticos, es cada vez más frecuente que los investigadores seleccionen la escala espacial más pequeña que permitan los datos. Sostenemos que esta práctica requiere un mayor escrutinio, ya que las escalas espaciales o temporales no necesariamente mejoran la calidad de las inferencias. Si bien los datos altamente desagregados reducen algunas amenazas para la inferencia (por ejemplo, el sesgo de agregación), aumentan el riesgo de otras (por ejemplo, la clasificación errónea de los resultados). Por lo tanto, sostenemos que los investigadores deberían adoptar un enfoque basándose más en principios a la hora de seleccionar la escala espacial para su análisis. Para contribuir a realizar esta elección, caracterizamos el problema de la agregación de los datos espaciales, analizamos las consecuencias de una agregación excesiva (o insuficiente) y ofrecemos algunas orientaciones para la investigación aplicada. Demostramos estas cuestiones utilizando tanto experimentos simulados como un análisis de los patrones de violencia en Afganistán. Les chercheurs ont maintenant un meilleur accès à des données granulaires géoréférencées (c-à-d, spatiales) sur les phénomènes politiques et sociaux que jamais auparavant. Ces données ont été largement utilisées, car elles offrent aux chercheurs le potentiel d’analyser des phénomènes locaux, de tester des mécanismes et de mieux comprendre les comportem","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"471 - 491"},"PeriodicalIF":1.3,"publicationDate":"2022-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48126664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Forecasting conflict in Africa with automated machine learning systems","authors":"Vito D'Orazio, Yu Lin","doi":"10.1080/03050629.2022.2017290","DOIUrl":"https://doi.org/10.1080/03050629.2022.2017290","url":null,"abstract":"Abstract The ViEWS problem is to forecast changes in the level of state-based violence for each of the next six months at the PRIO-GRID and country level. For this competition and toward the goal of improving sub-national and country level forecasts, we experiment with combinations of automated machine learning (autoML) systems and limited datasets that emphasize the endogenous nature of conflict. Two core findings emerge: autoML improves predictive performance and the Dynamics model performs best. The data used for the Dynamics model is limited to measures of state-based violence built from the event-level violence data plus those describing the spatial and temporal structure of the data. The intent is to capture spatial and temporal conflict dynamics while not overfitting to exogenous factors, which is especially problematic with flexible autoML algorithms and the types of highly disaggregate data used here. At the PGM level, this model won the ViEWS competition for “predictive accuracy” and split the win for “originality.” Beyond the ViEWS competition, we expect conflict forecasting models that couple advanced autoML systems with variables that reflect a diverse set of conflict dynamics to have high predictive performance, especially at sub-national and sub-annual aggregations. El problema del ViEWS es que predice los cambios en el nivel de violencia estatal de cada uno de los próximos seis meses a nivel de PRIO-GRID y de país. En el marco de esta competencia y con el objetivo de mejorar las predicciones a nivel regional y nacional, probamos combinaciones de sistemas de aprendizaje automático (autoML) y conjuntos de datos limitados que ponen de relieve la naturaleza endógena de los conflictos. Hay dos resultados principales: el autoML mejora el rendimiento predictivo y el modelo Dynamics es el que mejor funciona. Los datos utilizados para el modelo Dynamics se limitan a las medidas de la violencia a nivel estatal establecidas a partir de los datos de la violencia sobre eventos más los que describen la estructura espacial y temporal de los datos. La intención es captar la dinámica espacial y temporal de los conflictos sin caer en el exceso de ajuste de los factores exógenos, lo que supone un problema, sobre todo con los algoritmos autoML flexibles y los tipos de datos altamente desagregados que se utilizan aquí. A nivel de PGM, este modelo ganó la competencia del ViEWS tanto por su “precisión predictiva” como por su “originalidad”. Más allá de la competencia del ViEWS, esperamos que los modelos de previsión de conflictos que combinan sistemas avanzados de autoML con variables que reflejan un conjunto diverso de dinámicas de conflicto tengan un alto resultado predictivo, sobre todo en agregados regionales y semestrales. La problématique du ViEWS (Violence early-warning system, système d’alerte précoce sur la violence) est de prévoir les évolutions du niveau de violence étatique pour chacun des six prochains mois au niveau de la grille PRIO et au","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"714 - 738"},"PeriodicalIF":1.3,"publicationDate":"2022-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48606203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dictators, personalized security forces, and coups","authors":"Wonjun Song","doi":"10.1080/03050629.2021.1977638","DOIUrl":"https://doi.org/10.1080/03050629.2021.1977638","url":null,"abstract":"ABSTRACT Dictators rely on coercive forces to remain in office, as violence is the ultimate arbiter of power in these regimes. However, coercive forces also can remove the dictator from office in a coup. This presents the dictator with a dilemma. One way to address this dilemma is to personalize the security forces. This paper argues that personalizing the security forces decreases coup risk by: (a) linking the security elites’ fate more closely to the leader’s and (b) increasing the informational advantage the leader has over security elites. Using a new measure of the personalization of security apparatus, I show that personalization decreases coup risk in dictatorships, but this stabilizing effect of personalization disappears after the dictator’s exit from office. This study documents how dictators transform the security apparatus to stabilize their rule, with implications for how dictatorships survive and collapse.","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"204 - 232"},"PeriodicalIF":1.3,"publicationDate":"2022-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41771184","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recurrent neural networks for conflict forecasting","authors":"Iris Malone","doi":"10.1080/03050629.2022.2016736","DOIUrl":"https://doi.org/10.1080/03050629.2022.2016736","url":null,"abstract":"Abstract Can history predict the escalation of future violence? This research note evaluates the use of a Recurrent Neural Network (RNN) for the Violence Early Warning System (ViEWS) Prediction Competition. Existing research on civil conflict shows violence is a persistent and recurring process, often shaping the direction of future conflicts. Building on this insight, I build a RNN model to examine how well historical patterns in conflict predict long-term escalation trends. A RNN is a simple, but powerful machine learning tool for time series forecasting due to its capacity to learn long sequences of information. I show that an RNN model can produce relatively accurate forecasts due to systematic patterns in conflict processes, consistent with existing research on “conflict traps.” The results provide important lessons for conflict forecasting and ground opportunities for using RNN models in future political science research. ¿La historia puede predecir el aumento de la violencia en el futuro? Esta nota de investigación evalúa el uso de una red neuronal recurrente (Recurrent Neural Network, RNN) para la competencia de predicciones del Sistema de Alerta Temprana de Violencia (Violence Early Warning System, ViEWS). Las investigaciones existentes sobre los conflictos civiles demuestran que la violencia es un proceso persistente y recurrente que, a menudo, da forma a la dirección de futuros conflictos. Con base en esta percepción, elaboro un modelo de RNN para examinar la eficacia de los patrones históricos de los conflictos al momento de predecir tendencias a largo plazo. La RNN es una herramienta de aprendizaje automático sencilla, pero poderosa, para la predicción de series temporales debido a su capacidad para aprender secuencias extensas de información. Los resultados demuestran que el modelo genera pronósticos relativamente precisos en los Estados débiles y fallidos, lo cual coincide con las investigaciones existentes sobre las “trampas de conflictos.” No obstante, el modelo presenta dificultades para predecir nuevos conflictos civiles; esto es coincidente con las teorías informativas sobre el inicio de conflictos. Los resultados brindan lecciones importantes para la predicción de conflictos y demuestran oportunidades para las aplicaciones de RNN en futuras investigaciones sobre ciencias políticas. L’histoire peut-elle permettre de prédire l’escalade future de la violence ? Cet exposé de recherche évalue l’utilization d’un Réseau de neurones récurrents pour le concours de prédiction ViEWS (Violence early-warning system, système d’alerte précoce sur la violence). Des recherches existantes sur les conflits civils montrent que la violence est un processus persistant et récurrent qui façonne souvent l’orientation des conflits futurs. Je me suis appuyé sur cette idée pour développer un modèle de réseau de neurones récurrents dans l’objectif d’examiner à quel point les schémas historiques des conflits pouvaient permettre de prédire des tendances à ","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"614 - 632"},"PeriodicalIF":1.3,"publicationDate":"2022-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46156266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Covid-19 insecurities and migration aspirations","authors":"Miranda Simon, C. Schwartz, David Hudson","doi":"10.1080/03050629.2022.1991919","DOIUrl":"https://doi.org/10.1080/03050629.2022.1991919","url":null,"abstract":"ABSTRACT Using an original survey, this paper examines how pandemic-driven insecurities have affected aspirations to migrate internationally among youth in The Gambia. We find that individuals perceive wide inequalities between their government’s performance and the speed of Covid-19 recovery abroad. However, superior recovery abroad does not have significant effects on aspirations to migrate. Individual and local sources of security are more important: Individuals who were able to maintain their jobs throughout the pandemic are less likely to aspire to move abroad. The insecurity of Covid-19 job loss may be compensated by confidence in one’s government’s ability to tackle the pandemic. This suggests that, in the context of an event that has upended people’s lives, would-be migrants who managed to maintain a source of stability may seek comfort in familiar contexts; even if they appear worse than alternatives abroad.","PeriodicalId":51513,"journal":{"name":"International Interactions","volume":"48 1","pages":"309 - 326"},"PeriodicalIF":1.3,"publicationDate":"2022-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42871258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}