Publications

Taking time seriously: Predicting conflict fatalities using temporal fusion transformers

Conditionally accepted in International Interactions , 2026

In this paper we predict conflict fatalities with temporal fusion transformers as part of the 2023/24 edition of the VIEWS Prediction Challenge.

Recommended citation: Walterskirchen J, Oswald C, Häffner S, Binetti MN. Taking time seriously: Predicting conflict fatalities using temporal fusion transformers. Conditionally accepted, International Interactions https://doi.org/10.31235/osf.io/7xu93_v2

Text as Data for Crisis-Early Warning: A Comparative Assessment of NLP Methods for Conflict Prediction

Accepted in Conflict Management and Peace Science , 2026

This paper evaluates the performance of features extracted from a conflict dictionary, two sentiment dictionaries, a word-scaling approach, dynamic topic models, and a transformer model on a classical conflict prediction task.

Recommended citation: Walterskirchen, J. (2026). Text as Data for Crisis-Early Warning: A Comparative Assessment of NLP Methods for Conflict Prediction. Conflict Management and Peace Science https://doi.org/10.1177/07388942261422045

Computational and Robustness Reproducibility of UN Peacekeeping and Democratization in Conflict-Affected Countries

Replication report in EconStor , 2024

As part of the Replication Games (I4R) we replicated UN Peacekeeping and Democratization in Conflict-Affected Countries.

Recommended citation: Oswald, C., Walterskirchen, J. (2023, July 5). Computational and Robustness Reproducibility of UN Peacekeeping and Democratization in Conflict-Affected Countries. https://hdl.handle.net/10419/301427 https://hdl.handle.net/10419/301427

Replication of The Morning After: Report from the Nottingham Replication Games

Replication report in SocArXiv , 2023

As part of the Nottingham Replication Games (I4R) we replicated “The morning after: cabinet instability and the purging of ministers after failed coup attempts in autocracies”.

Recommended citation: Oswald, C., Walterskirchen, J., Häffner, S., Binetti, M. N., & Dworschak, C. (2023, July 5). Replication of The Morning After: Report from the Nottingham Replication Games. https://doi.org/10.31235/osf.io/a2d5p https://doi.org/10.31235/osf.io/a2d5p

Introducing an Interpretable Deep Learning Approach to Domain-Specific Dictionary Creation: A Use Case for Conflict Prediction

Published in Political Analysis, 2023

In this paper, we introduce an interpretable deep learning approach to automatic dictionary creation for conflict research.

Recommended citation: Häffner, S., Hofer, M., Nagl, M., & Walterskirchen, J. (2023). Introducing an Interpretable Deep Learning Approach to Domain-Specific Dictionary Creation: A Use Case for Conflict Prediction. Political Analysis, 31(4), 481-499. doi:10.1017/pan.2023.7 https://doi.org/10.1017/pan.2023.7

Sanction Dynamics in the Cases of North Korea, Iran, and Russia

Published in SpringerBriefs in International Relations, 2022

In this book, we analyse the complete life cycle of sanctions in three case studies.

Recommended citation: Walterskirchen, J., Mangott, G., Wend, C. (2022). Sanction Dynamics in the Cases of North Korea, Iran, and Russia: Objectives, Measures and Effects. SpringerBriefs in International Relations. doi:10.1007/978-3-031-17397-4 https://doi.org/10.1007/978-3-031-17397-4

A Cross-National Analysis of Forced Population Resettlement in Counterinsurgency Campaigns

Published in Peace Economics, Peace Science and Public Policy, 2019

In this paper we study the use of Forced Population Resettlement in Counterinsurgency Campaigns.

Recommended citation: Böhmelt, Tobias, Dworschak, Christoph, Pilster, Ulrich and Walterskirchen, Julian. "A Cross-National Analysis of Forced Population Resettlement in Counterinsurgency Campaigns" Peace Economics, Peace Science and Public Policy, vol. 26, no. 1, 2020, pp. 20190022. https://doi.org/10.1515/peps-2019-0022 https://doi.org/10.1515/peps-2019-0022

Julian Walterskirchen