{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T02:17:03Z","timestamp":1777861023453,"version":"3.51.4"},"reference-count":57,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T00:00:00Z","timestamp":1775779200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T00:00:00Z","timestamp":1775779200000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Expert Systems"],"published-print":{"date-parts":[[2026,5]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>\n                    The accelerating infusion of advanced computational methods into geopolitical analysis has created new opportunities to anticipate unrest, economic shocks and diplomatic shifts. Traditional machine learning pipelines can extract statistical patterns from large event corpora, but they often struggle to incorporate real\u2010time contextual information or explain their predictions in language accessible to decision\u2010makers. This study proposes a comprehensive framework,\n                    <jats:sc>LLM4Geopolitics<\/jats:sc>\n                    , that couples a domain\u2010adapted large language model with a retrieval\u2010augmented generation mechanism grounded in a structured knowledge graph. The forecasting component employs a transformer architecture tailored to sparse, irregular event streams, while the generative component translates model outputs into dialogue\u2010ready assessments enriched with up\u2010to\u2010date economic and peace\u2010index indicators. Experiments conducted on the\n                    <jats:sc>Gdelt<\/jats:sc>\n                    dataset demonstrate that the integrated approach improves event\u2010severity prediction and generates fact\u2010consistent narratives compared with baseline time series and text\u2010only models. These findings highlight the potential of combining specialised sequence models, on\u2010demand knowledge retrieval and generative reasoning to deliver timely and interpretable insights for geopolitical forecasting.\n                  <\/jats:p>","DOI":"10.1111\/exsy.70258","type":"journal-article","created":{"date-parts":[[2026,4,11]],"date-time":"2026-04-11T03:06:15Z","timestamp":1775876775000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["<scp>LLM4Geopolitics<\/scp>\n                    : A Framework Leveraging Large Language Models for Predicting Geopolitical Events"],"prefix":"10.1111","volume":"43","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1346-3851","authenticated-orcid":false,"given":"Amira","family":"Mouakher","sequence":"first","affiliation":[{"name":"Espace Dev UMR 228 UPVD, IRD, UM, UG University of Perpignan  Perpignan France"},{"name":"Future Potentials Observatory, MOME Foundation  Budapest Hungary"}]},{"given":"Nuno","family":"Morgado","sequence":"additional","affiliation":[{"name":"Future Potentials Observatory, MOME Foundation  Budapest Hungary"},{"name":"CIAS, Corvinus University of Budapest  Budapest Hungary"}]},{"given":"Farah","family":"Ftouhi","sequence":"additional","affiliation":[{"name":"School of Computer Science University of Birmingham  Birmingham UK"}]}],"member":"311","published-online":{"date-parts":[[2026,4,10]]},"reference":[{"key":"e_1_2_12_2_1","unstructured":"Armant V. A.Mouakher F.Vargas\u2010Rojas et\u00a0al.\u201cCan Knowledge Graphs and Retrieval\u2010Augmented Generation be Combined to Explain Query\/Answer Relationships Truthfully?\u201din DAO\u2010XAI 2024 Data meets Ontologies in Explainable AI co\u2010located with the 27th European Conference on Artificial Intelligence (ECAI 2024) 3833 2024."},{"key":"e_1_2_12_3_1","doi-asserted-by":"publisher","DOI":"10.1080\/19331681.2024.2417263"},{"key":"e_1_2_12_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.procs.2024.09.576"},{"key":"e_1_2_12_5_1","unstructured":"Atlantic Council.\u201cLeveraging Generative Artificial Intelligence to Outcompete Strategic Rivals.\u201dAccessed April 14 2025.https:\/\/www.atlanticcouncil.org\/in\u2010depth\u2010research\u2010reports\/issue\u2010brief\/leveraging\u2010generative\u2010artificial\u2010intelligence\u2010to\u2010outcompete\u2010strategic\u2010rivals\/. issue brief (2024)."},{"key":"e_1_2_12_6_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0254350"},{"key":"e_1_2_12_7_1","doi-asserted-by":"crossref","unstructured":"Bajak F.\u201cUs Intelligence Agencies' Embrace of Generative AI Is at Once Wary and Urgent.\u201dPBS News May 24 2024.","DOI":"10.1201\/9781003565703-2"},{"key":"e_1_2_12_8_1","doi-asserted-by":"crossref","unstructured":"Baytas I. M. C.Xiao X.Zhang F.Wang A. K.Jain andJ.Zhou.2017.\u201cPatient Subtyping Via Time\u2010Aware LSTM Networks.\u201din Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 65\u201374.","DOI":"10.1145\/3097983.3097997"},{"key":"e_1_2_12_9_1","doi-asserted-by":"publisher","DOI":"10.1002\/aaai.12085"},{"key":"e_1_2_12_10_1","unstructured":"Bommasani R. D. A.Hudson E.Adeli et\u00a0al.\u201cOn the Opportunities and Risks of Foundation Models.\u201dpreprint arXiv 2021 2108.07258."},{"key":"e_1_2_12_11_1","first-page":"1877","article-title":"Language Models Are Few\u2010Shot Learners","volume":"33","author":"Brown T.","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_12_12_1","doi-asserted-by":"publisher","DOI":"10.1177\/0022343313507302"},{"key":"e_1_2_12_13_1","doi-asserted-by":"publisher","DOI":"10.1613\/jair.953"},{"key":"e_1_2_12_14_1","unstructured":"Chung J. C.Gulcehre K.Cho andY.Bengio.\u201cEmpirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling.\u201dpreprint arXiv 2014 1412.3555 1412."},{"key":"e_1_2_12_15_1","volume-title":"Geopolitics: The Geography of International Relations","author":"Cohen S. B.","year":"2014"},{"key":"e_1_2_12_16_1","doi-asserted-by":"publisher","DOI":"10.1111\/exsy.13667"},{"key":"e_1_2_12_17_1","unstructured":"Croicu M. andS. P.von derMaase.2025.\u201cFrom Newswire to Nexus: Using Text\u2010Based Actor Embeddings and Transformer Networks to Forecast Conflict Dynamics.\u201darXiv preprint arXiv:2501.03928."},{"key":"e_1_2_12_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/OJCOMS.2025.3573194"},{"issue":"1","key":"e_1_2_12_19_1","first-page":"1","article-title":"Large Language Models Empowered Agent\u2010Based Modeling and Simulation: A Survey and Perspectives","volume":"11","author":"Gao C.","year":"2023","journal-title":"AI Communications"},{"key":"e_1_2_12_20_1","unstructured":"Ghosh S. C. K. R.Evuru S.Kumar et\u00a0al.\u201cA Closer Look at the Limitations of Instruction Tuning.\u201dpreprint arXiv 2024 2402.05119."},{"key":"e_1_2_12_21_1","unstructured":"Gu J. X.Jiang Z.Shi et\u00a0al.\u201cA Survey on LLM\u2010as\u2010a\u2010Judge.\u201dpreprint arXiv 2024 2411.15594."},{"key":"e_1_2_12_22_1","doi-asserted-by":"crossref","unstructured":"Halterman A.\u201cLinking Events and Locations in Political Text.\u201dMIT Political Science Department Research Paper No. 2018\u201021 2018.10.2139\/ssrn.3267476.","DOI":"10.2139\/ssrn.3267476"},{"key":"e_1_2_12_23_1","unstructured":"Halterman A.\u201cExtracting Political Events From Text Using Syntax and Semantics.\u201dWorking Paper 2020.https:\/\/andrewhalterman.com\/files\/Halterman_extracting_events.pdf."},{"key":"e_1_2_12_24_1","doi-asserted-by":"publisher","DOI":"10.1177\/0022343319823860"},{"key":"e_1_2_12_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.scs.2022.104026"},{"issue":"2","key":"e_1_2_12_26_1","first-page":"3","article-title":"Lora: Low\u2010Rank Adaptation of Large Language Models","volume":"1","author":"Hu E. J.","year":"2022","journal-title":"ICLR"},{"key":"e_1_2_12_27_1","doi-asserted-by":"crossref","unstructured":"Huber D. J. S.Johnson N.Stepp et\u00a0al.2019.\u201cMatrics: A System for Human\u2010Machine Hybrid Forecasting of Geopolitical Events.\u201din Proceedings of the IEEE International Conference on Big Data (Big Data) IEEE 2028\u20132032.","DOI":"10.1109\/BigData47090.2019.9006434"},{"key":"e_1_2_12_28_1","volume-title":"A Simple Guide to Retrieval Augmented Generation","author":"Kimothi A.","year":"2025"},{"key":"e_1_2_12_29_1","doi-asserted-by":"publisher","DOI":"10.1177\/2053951714528481"},{"key":"e_1_2_12_30_1","first-page":"9459","article-title":"Retrieval\u2010Augmented Generation for Knowledge\u2010Intensive Nlp Tasks","volume":"33","author":"Lewis P.","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_12_31_1","doi-asserted-by":"crossref","unstructured":"Liao R. X.Jia Y.Li Y.Ma andV.Tresp.\u201cGENTKG: Generative Forecasting on Temporal Knowledge Graph With Large Language Models.\u201din Findings of the associat1ion for computational linguistics: NAACL 2024 4303\u20134317.","DOI":"10.18653\/v1\/2024.findings-naacl.268"},{"key":"e_1_2_12_32_1","doi-asserted-by":"publisher","DOI":"10.1177\/00027642211021650"},{"key":"e_1_2_12_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2024.124456"},{"key":"e_1_2_12_34_1","doi-asserted-by":"publisher","DOI":"10.1080\/00396338.2022.2032959"},{"key":"e_1_2_12_35_1","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511809071"},{"key":"e_1_2_12_36_1","doi-asserted-by":"publisher","DOI":"10.48088\/ejg.n.mor.14.4.013.021"},{"key":"e_1_2_12_37_1","doi-asserted-by":"publisher","DOI":"10.30965\/24518921\u201012341045"},{"key":"e_1_2_12_38_1","doi-asserted-by":"publisher","DOI":"10.1017\/dap.2024.27"},{"key":"e_1_2_12_39_1","unstructured":"Nemkova P. A. S. O.Polat R. I.Jahan et\u00a0al.\u201cTowards Automated Situation Awareness: A Rag\u2010Based Framework for Peacebuilding Reports.\u201dpreprint arXiv 2025 2505.10586."},{"key":"e_1_2_12_40_1","unstructured":"Nie Y. N. H.Nguyen P.Sinthong andJ.Kalagnanam.2023.\u201cA Time Series Is Worth 64 Words: Long\u2010Term Forecasting With Transformers.\u201din International Conference on Learning Representations (ICLR)."},{"key":"e_1_2_12_41_1","doi-asserted-by":"crossref","unstructured":"Redl C. andS.Hlatshwayo.\u201cForecasting Social Unrest: A Machine Learning Approach.\u201dIMF Working Papers 2021\/263 International Monetary Fund (2021).","DOI":"10.5089\/9781557758873.001"},{"key":"e_1_2_12_42_1","doi-asserted-by":"publisher","DOI":"10.1111\/exsy.13763"},{"key":"e_1_2_12_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.techsoc.2023.102372"},{"key":"e_1_2_12_44_1","unstructured":"Shi X. S.Xue K.Wang et\u00a0al.2023.\u201cLanguage Models Can Improve Event Prediction by Few\u2010Shot Abductive Reasoning.\u201darXiv preprint arXiv:2305.16646.https:\/\/arxiv.org\/abs\/2305.16646."},{"key":"e_1_2_12_45_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.physa.2024.129977"},{"key":"e_1_2_12_46_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ipm.2009.03.002"},{"key":"e_1_2_12_47_1","doi-asserted-by":"publisher","DOI":"10.1111\/exsy.13531"},{"key":"e_1_2_12_48_1","doi-asserted-by":"publisher","DOI":"10.1198\/073500102317351921"},{"key":"e_1_2_12_49_1","unstructured":"Stoehr N. L. T.Hennigen J.Valvoda R.West R.Cotterell andA.Schein.\u201cAn Ordinal Latent Variable Model of Conflict Intensity.\u201dpreprint arXiv 2022 2210.03971."},{"key":"e_1_2_12_50_1","unstructured":"Su J. C.Jiang X.Jin et\u00a0al.\u201cLarge Language Models for Forecasting and Anomaly Detection: A Systematic Literature Review.\u201dpreprint arXiv 2024 2402.10350.https:\/\/arxiv.org\/abs\/2402.10350."},{"key":"e_1_2_12_51_1","doi-asserted-by":"publisher","DOI":"10.1145\/3516367"},{"key":"e_1_2_12_52_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.techsoc.2024.102739"},{"key":"e_1_2_12_53_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11229-022-03485-5"},{"key":"e_1_2_12_54_1","unstructured":"Wu H. T.Hu Y.Liu H.Zhou J.Wang andM.Long.2023.\u201cTimesnet: Temporal 2d\u2010Variation Modeling for General Time Series Analysis.\u201din: International Conference on Learning Representations (ICLR)."},{"key":"e_1_2_12_55_1","doi-asserted-by":"crossref","unstructured":"Yu C. Z.Weng Y.Li Z.Li X.Hu andY.Zhao.\u201cTowards More Accurate US Presidential Election via Multi\u2010Step Reasoning With Large Language Models.\u201dpreprint arXiv 2025 2411.03321.","DOI":"10.2139\/ssrn.4996124"},{"key":"e_1_2_12_56_1","unstructured":"Zhang L. andY.Ning.\u201cLarge Language Models as Interpolated and Extrapolated Event Predictors.\u201dpreprint arXiv 2024 2406.10492.https:\/\/arxiv.org\/abs\/2406.10492."},{"key":"e_1_2_12_57_1","doi-asserted-by":"crossref","unstructured":"Zhou H. S.Zhang J.Peng et\u00a0al.2021.\u201cInformer: Beyond Efficient Eransformer for Long Aequence Time\u2010Series Forecasting.\u201din Proceedings of the AAAI Conference on Artificial Intelligence 35 11106\u201311115.","DOI":"10.1609\/aaai.v35i12.17325"},{"key":"e_1_2_12_58_1","first-page":"27293","article-title":"Forecasting Future World Events With Neural Networks","volume":"35","author":"Zou A.","year":"2022","journal-title":"Advances in Neural Information Processing Systems"}],"container-title":["Expert Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/exsy.70258","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/exsy.70258","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/exsy.70258","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T02:43:27Z","timestamp":1777517007000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/exsy.70258"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,10]]},"references-count":57,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,5]]}},"alternative-id":["10.1111\/exsy.70258"],"URL":"https:\/\/doi.org\/10.1111\/exsy.70258","archive":["Portico"],"relation":{},"ISSN":["0266-4720","1468-0394"],"issn-type":[{"value":"0266-4720","type":"print"},{"value":"1468-0394","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,10]]},"assertion":[{"value":"2024-10-21","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-30","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-10","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70258"}}