Elamur Research⌕
☰
home / papers / 2026 / GENOME: A New Geopolitical Event Methodology and Dataset using Large Language Models

GENOME: A New Geopolitical Event Methodology and Dataset using Large Language Models

Paper · 2026 · Alessandro Dell'Orto and Jesse Kommandeur

Summary

Event data on international relations depends on automated coding. POLECAT stopped updating around August 2024, which left no current machine-coded dataset that tracks both hostile and cooperative interactions between states. The authors address this with GENOME, a dataset built from English newswire using an LLM pipeline. It follows the 16 event types of the PLOVER ontology and adds a Third Party role to the usual Actor–Recipient pair, so that entities which give context to multilateral events have a place. The pipeline has four stages: extraction, classification, entity normalization and deduplication. It uses OpenAI GPT models with one-shot prompts and schema-constrained outputs. No gold standard exists, so the authors compare GENOME with POLECAT over five shared months instead. The two agree well on conflict events. GENOME, however, records many more cooperative and especially verbal events, dates events by when they happened rather than when they were reported, and separates international organizations at a finer level. The work matters because it restores an openly released, continuously updated resource for conflict early-warning research.

Key points

  • Corpus: about 148,000 English articles from roughly 2,400 sources, covering January 2024 to January 2026. The comparison with POLECAT uses February–June 2024 (about 28,000 events).
  • Models: gpt-4o-mini summarizes the first ~200 words of each article and infers the event date. A fine-tuned gpt-4.1-nano assigns PLOVER classes; its training used silver labels produced by gpt-5.1.
  • Entity normalization:
    • Names are fuzzy-matched against 248 countries and 40 organizations, which cut unique country/IO entries by 81.7%.
    • Sentence-Transformer embeddings clustered with FAISS reduced agent names by 35.4%, to 32,624.
    • GENOME distinguishes bodies such as NATO, IMF, WTO, BRICS and ICC.
  • Limitations:
    • Because all sources come from a single English-language provider, the Caucasus, Latin America and South/Southeast Asia are only thinly represented.
    • There is no gold standard for computing precision, recall or F1.
    • 4.4% of events have negative date lags.
    • The deduplication threshold (0.80) and the clustering steps have not been checked against human judgments.
  • Release: the data and analysis scripts are available at github.com/HCSS-Data-Lab/Submission-GENOME.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

Also on the byline: The Hague Centre for Strategic Studies.

 Fields, topics and concepts

📚 Shelves 7The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksSocial Sciences · 1,772 worksGeneral Social Sciences · 227 worksPolitical Science and International Relations · 182 works

Cite

Cite as Alessandro Dell'Orto and Jesse Kommandeur. 2026. GENOME: A New Geopolitical Event Methodology and Dataset using Large Language Models. Proceedings of the 9th Workshop on Event Extraction and Understanding: Challenges and Applications (EEUCA 2026).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.eeuca-1.9

@inproceedings{dellorto2026genome,
  title = {{GENOME: A New Geopolitical Event Methodology and Dataset using Large Language Models}},
  author = {Alessandro Dell'Orto and Jesse Kommandeur},
  booktitle = {Proceedings of the 9th Workshop on Event Extraction and Understanding: Challenges and Applications (EEUCA 2026)},
  year = {2026},
  doi = {10.18653/v1/2026.eeuca-1.9},
  url = {https://doi.org/10.18653/v1/2026.eeuca-1.9}
}

References

32 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Kaiqian Ahrabian, Ethan Boxer, and Jay Pujara. 2025. Toward better temporal structures for geopolitical events forecasting. In Proceedings of the 4th Workshop on Generation, Evaluation & Metrics (GEM 2024), pages 588–602. Association for Computational Linguistics. find
  2. Kiymet Akdemir and Ali Hürriyetoğlu. 2022. Zero-shot ranking socio-political texts with transformer language models to reduce close reading time. In Proceedings of the 5th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2022), pages 124–132.… find
  3. Edward E. Azar. 1980. The conflict and peace data bank (COPDAB) project. Journal of Conflict Resolution, 24(1):143–152. find
  4. Rui Bai, Dong Lu, Sheng Ran, Emily M. Olson, Himank Lamba, Allison Cahill, Joel Tetreault, and Alejandro Jaimes. 2025. CEHA: A dataset of conflict events in the horn of africa. In Proceedings of the 31st International Conference on Computational Linguistics (COLING 2025). Association for Computation… find
  5. Kyle Beardsley, Patrick James, Jonathan Wilkenfeld, and David Quinn. 2024. What is escalation? Measuring crisis dynamics in international relations with human and LLM generated event data. ArXiv:2402.03340. — [arXiv:2402.03340](https://arxiv.org/abs/2402.03340) find
  6. Doug Bond, Brian Bennett, and William Vogele. 1994. Data development and interaction events analysis using KEDS/PANDA: An interim report. Technical report, International Studies Association, Washington, DC. find
  7. Kalev Leetaru and Philip A. Schrodt. 2013. GDELT: Global data on events, location, and tone, 1979–2012. In ISA Annual Convention, volume 2, pages 1–49. find
  8. Charles A. McClelland. 1961. The acute international crisis. World Politics, 14(1):182–204. find
  9. Patrick T. Brandt and Marlo Sianan. 2025. Measurement of event data from text. Frontiers in Political Science, 6. find
  10. Charles A. McClelland. 1978. World event/interaction survey, 1966–1978. Technical Report 5211, ICPSR. find
  11. Erica Cai and Brendan O’Connor. 2023. A monte carlo language model pipeline for zero-shot sociopolitical event extraction. ArXiv:2305.15051. — [arXiv:2305.15051](https://arxiv.org/abs/2305.15051) find
  12. Shrey Meher and Patrick T. Brandt. 2025. ConflLlama: Domain-specific adaptation of large language models for conflict event classification. Research & Politics, 12(3). find
  13. Ruoxi Chen, Chengzhi Qin, Wenxing Jiang, and Donna Choi. 2024. Is a large language model a good annotator for event extraction? In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 17772–17780. find
  14. Sean P. O’Brien. 2010. Crisis early warning and decision support: Contemporary approaches and thoughts on future research. International Studies Review, 12(1):87–104. find
  15. Helene Brinken Olsen, Étienne Simon, Erik Velldal, and Lilja Øvrelid. 2024. Socio-political events of conflict and unrest: A survey of available datasets. In Proceedings of the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2024), pages… find
  16. Jun Gao, Huan Zhao, Changlong Yu, and Ruifeng Xu. 2023. Exploring the feasibility of ChatGPT for event extraction. Preprint, arXiv:2303.03836.
  17. Deborah J. Gerner, Philip A. Schrodt, Omür Yilmaz, and Rajaa Abu-Jabr. 2002. Conflict and mediation event observations (CAMEO): A new event data framework for the analysis of foreign policy interactions. Paper presented at the International Studies Association, New Orleans. find
  18. Open Event Data Alliance. 2024. PLOVER: Political language ontology for verifiable event records — event, actor and data interchange specification (draft version 2.0). find
  19. Myrsini Halkia, Stefano Ferri, Michail Papazoglou, and Marlies Kampen Schellens. 2020. Dynamic global conflict risk index. Technical report, Publications Office of the European Union. JRC Technical Report. find
  20. Clionadh Raleigh, Roudabeh Kishi, and Andrew Linke. 2023. Political instability patterns are obscured by conflict dataset scope conditions, sources, and coding choices. Humanities and Social Sciences Communications, 10:74. find
  21. Andrew Halterman, Benjamin E. Bagozzi, Andreas Beger, Philip A. Schrodt, and Grace I. Scarborough. 2023a. PLOVER and POLECAT: A new political event ontology and dataset. SocArXiv. find
  22. Shehzad Salam, Patrick Brandt, Vito D’Orazio, Jennifer Holmes, Javier Osorio, and Latifur Khan. 2020. An online structured political event dataset based on CAMEO ontology. SocArXiv. find
  23. Andrew Halterman and Katherine A. Keith. 2025. Codebook LLMs: Evaluating LLMs as measurement tools for political science concepts. Political Analysis. find
  24. Grace I. Scarborough, Benjamin E. Bagozzi, Andreas Beger, John Berrie, Andrew Halterman, Philip A. Schrodt, and Jevon Spivey. 2023. POLECAT weekly data. find
  25. Andrew Halterman, Philip A. Schrodt, Andreas Beger, Benjamin E. Bagozzi, and Grace I. Scarborough. 2023b. Creating custom event data without dictionaries: A bag-of-tricks. ArXiv:2304.01331. — [arXiv:2304.01331](https://arxiv.org/abs/2304.01331) find
  26. Sina J. Semnani, Peng Zhang, Wenxuan Zhai, Hao Li, Nicholas Beauchamp, Tyler Billing, Koko Kishi, Michaela Li, and Monica Lam. 2025. LEMON-ADE: A large multilingual expert-annotated abstractive event dataset for the real world. In Findings of the Association for Computational Linguistics: ACL 2025,… find
  27. J. Craig Jenkins, Charles L. Taylor, Marianne Abbott, Thomas V. Maher, and Lindsey Peterson. 2012. The world handbook of political indicators IV. Technical report, Mershon Center for International Security Studies, The Ohio State University. find
  28. Gary King and Will Lowe. 2003a. 10 million international dyadic events. IQSS Dataverse Network. find
  29. Surendrabikram Thapa, Suresh Adhikari, Hristo Tanev, and Ali Hürriyetoğlu. 2025. Challenges and applications of automated extraction of socio-political events at the age of large language models. In Proceedings of the 8th Workshop on Challenges and Applications of Automated Extraction of Socio-poli… find
  30. Gary King and Will Lowe. 2003b. An automated information extraction tool for international conflict data with performance as good as human coders: A rare events evaluation design. International Organization, 57(3):617–642. find
  31. Jay Yonamine. 2016. A guide to event data: Past, present, and future. All Azimuth: A Journal of Foreign Policy and Peace. find
  32. Huiling You, Samia Touileb, Lilja Øvrelid, and Erik Velldal. 2025. Event-based evaluation of abstractive news summarization. In Proceedings of the Fourth Workshop on Generation, Evaluation and Metrics (GEM 2 ), pages 504–510. Association for Computational Linguistics. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/orto-2026-genome-new-geopolitical-event-methodology/