Author(s)

ZhiMin Liu^*, QingHua Zhou

Affiliation(s)

College of Systems Engineering, National University of Defense Technology, Changsha 410073, Hunan, China.

Corresponding Author

ZhiMin Liu

ABSTRACT

The scientific rigour and precision of major strategic decisions at the national level have been demonstrated to directly impact a nation's sovereignty, economic security, and diplomatic standing. In the contemporary geopolitical landscape, characterised by the proliferation of intricate international relations, the discord between established systems and emergent demands has attained a heightened degree of visibility. This has resulted in a substantial escalation in the intricacy and ambiguity of strategic conflicts among nations. Among the various measures employed, the implementation of sanction policies directed against specific countries is of particular significance in achieving political and national security objectives. However, the factors that influence sanction policies are characterised by their multi-dimensionality and strong interdependencies (highly coupled nature). Conventional research paradigms, predicated on qualitative analysis, are progressively inadequate to satisfy the decision-making demands inherent in complex scenarios. This has given rise to a pressing need for AI technologies to construct efficient and precise analytical frameworks. The present paper proposes a deep neural network-based analytical method for the evaluation of sanction policies enacted by nations (or international organizations). The efficacy of this method is demonstrated by its ability to predict the potential outcomes of such policies, thereby providing critical information and knowledge support for the scientific formulation of national sanction strategies. The core approach involves the curation of data from extensive historical records to form quantitative indicators related to sanction policies. A deep neural network is then employed to model the intrinsic relationships between these indicators and the effects of sanctions, enabling the prediction of outcomes in new contexts and offering decision support for critical policy formulation. In conclusion, the model was trained on the Global Sanctions Database (GSDB-R3), resulting in an efficient predictive model for sanction policy effectiveness. The analysis of metrics such as accuracy and recall demonstrates the feasibility and effectiveness of the proposed method in predicting the outcomes of sanction policies.

KEYWORDS

Artificial neural network; Strategic sanctions; Predictive analysis

CITE THIS PAPER

ZhiMin Liu, QingHua Zhou. A data-driven model and methodology for predicting the efficacy of strategic sanctions. Trends in Social Sciences and Humanities Research. 2025, 3(6): 24-34. DOI: https://doi.org/10.61784/tsshr3179.

REFERENCES

[1] Ozgür Ozdamar, Evgeniia Shahin. Consequences of Economic Sanctions: The State of the Art and Paths Forward. International Studies Review, 2021, 23: 1646-1671.

[2] Kavakli, Kerim Can, Chatagnier, et al. The Power to Hurt and the Effectiveness of International Sanctions. The Journal of Politics, 2020, 82(3): 879-894. 

[3] Katharina L Meissner, Patrick A Mello. The unintended consequences of UN sanctions: A qualitative comparative analysis. Contemporary Security Policy, 2022, 43: 243-273.

[4] Meng Xiaoyu, Wang Zhongmin, Jing Shenqi, et al. The development and challenge of medical artificial intelligence. China Digital Medicine, 2019, 14(03): 15-17.

[5] Wang J D, Yang J, Pei Y Z, et al. Research on distribution network fault aided decision based on knowledge graph. Power Grid Technology, 2021, 45(06): 2101-2112.

[6] Chen Yao. Research on Intelligent Maintenance Decision Support System of Expressway. Traffic World, 2020, 538(16): 148-149.

[7] Felbermayr G J, Kirilakha I, Prat J. The Global Sanctions Data Base – Release 3: COVID-19, Russia, and Multilateral Sanctions. CESifo Working Paper Series No. 10101, 2022. https://cesifo.org/DocDL/cesifo1_wp10101.pdf.

[8] Felbermayr G J, Lein J, Prat J. The Global Sanctions Data Base. CESifo Working Paper Series No. 8439, 2020. https://cesifo.org/DocDL/cesifo1_wp8439.pdf.