This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.
Mayank Garg*,
- Student, Department of AI-DS, Dr. Akhilesh Das Gupta Institute of Professional Studies (ADGIPS) Delhi,, New Delhi, India
Abstract
Military expenditure analysis has emerged as a critical component of economic and geopolitical intelligence in the modern era. This paper presents CIPHER Intelligence, a comprehensive AI-powered platform for analyzing and predicting global military spending patterns across 211 countries spanning
54 years (1970-2024). We employ advanced machine learning techniques, particularly Random Forest regression models, to achieve 99.5% prediction accuracy for military expenditure forecasting based on economic indicators. The platform integrates data from multiple authoritative sources including SIPRI, World Bank, and IMF databases, processing over 10,761 country- year observations. Our methodology encompasses comprehensive data preprocessing, feature engineering, and multi-dimensional analysis including regional, temporal, and economic development
perspectives. The system achieves remarkable predictive per- formance with R2 scores of 0.995 for absolute spending, 0.832 for GDP percentage, and 0.980 for per capita expenditure. Key findings reveal strong correlations between GDP size and military spending (r = 0.94), significant regional disparities in defense burden, and accelerating global military expenditure reaching
$2.4 trillion in 2024. The platform’s web-based interface provides real-time predictions, interactive visualizations, and comprehen- sive analytical dashboards for researchers, policymakers, and defense analysts. This work demonstrates the effectiveness of machine learning in economic intelligence analysis and establishes a robust framework for predictive defense economics.
Keywords: military expenditure, machine learning, predic- tive analytics, economic intelligence, defense economics, Random Forest, data analytics, SIPRI, geopolitical analysis
[This article belongs to Research & Reviews : Journal of Statistics ]
Mayank Garg*. CIPHER Intelligence: AI-Powered Global Military Expenditure Analysis and Predictive Modeling. Research & Reviews : Journal of Statistics. 2026; 15(01):-.
Mayank Garg*. CIPHER Intelligence: AI-Powered Global Military Expenditure Analysis and Predictive Modeling. Research & Reviews : Journal of Statistics. 2026; 15(01):-. Available from: https://journals.stmjournals.com/rrjost/article=2026/view=241856
References
[1] Stockholm International Peace Research Institute, “SIPRI Military Expenditure Database,” 2024. [Online]. Available: https://www.sipri.org/databases/milex
[2] World Bank, “World Development Indicators,” 2024. [Online]. Avail- able: https://databank.worldbank.org
[3] T. Sandler and K. Hartley, “Economics of defense: A survey,” Economic Journal, vol. 45, no. 3, pp. 1039-1066, 1995.
[4] P. Dunne and E. Sko¨ns, “Military expenditure and economic growth,”
Defence and Peace Economics, vol. 27, no. 5, pp. 599-613, 2016.
[5] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5-32, 2001.
[6] J. Friedman, “Greedy function approximation: A gradient boosting machine,” Annals of Statistics, vol. 29, no. 5, pp. 1189-1232, 2001.
[7] C. Nordhaus, D. Oneal, and B. Russett, “The effects of interstate system on military expenditures,” International Studies Quarterly, vol. 56, no. 3, pp. 391-402, 2012.
[8] D. Hewitt, “Military expenditure worldwide: Determinants and trends, 1972-1988,” Journal of Public Policy, vol. 12, no. 2, pp. 105-152, 1992.
[9] N. Halko, P. Martinsson, and J. Tropp, “Finding structure with ran- domness: Probabilistic algorithms for constructing approximate matrix decompositions,” SIAM Review, vol. 53, no. 2, pp. 217-288, 2011.
[10] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521,
pp. 436-444, 2015.
[11] F. Pedregosa et al., “Scikit-learn: Machine learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825-2830, 2011.
[12] S. Heston and S. Sinha, “News vs. sentiment: Predicting stock returns from news stories,” Financial Analysts Journal, vol. 73, no. 3, pp. 67-83, 2017.
Research & Reviews : Journal of Statistics
| Volume | 15 |
| Issue | 01 |
| Received | 07/04/2026 |
| Accepted | 17/04/2026 |
| Published | 29/04/2026 |
| Publication Time | 22 Days |
Login
PlumX Metrics