The escalating global digitization has positioned cybercrime as a critical threat to socio-economic stability, with estimated global costs exceeding $8 trillion annually. This study presents a comprehensive quantitative analysis of global cybercrime distribution patterns and their key drivers through an integrated multi-model approach. We constructed a sophisticated evaluation framework encompassing five critical dimensions: legal infrastructure, technological capability, organizational maturity, capacity building, and international cooperation. Utilizing data from 150 countries spanning 2018-2023, we implemented an entropy-weighted WAM model to determine objective indicator weights, followed by a K-means clustering algorithm for country classification. Furthermore, we developed an advanced multiple linear regression model incorporating dynamic lag effects to assess cybersecurity policy effectiveness, complemented by an XGBoost model with SHAP analysis for demographic correlation mapping. Our results reveal four distinct national clusters, including high-success-rate nations (e.g., Myanmar, Cambodia) demonstrating 67% higher vulnerability rates, and high-prevention-capability nations (e.g., Denmark, Germany) showing 89% higher threat mitigation efficiency. The research confirms significant time-lagged policy impacts, with technology investments showing 45% greater effectiveness after 18-24 months. Demographic analysis establishes strong positive correlations between cybercrime density and internet penetration (r=0.82, p<0.01), while revealing negative correlations with education expenditure (r=-0.71, p<0.01). This research provides an evidence-based framework for developing targeted cybersecurity policies and resource allocation strategies.

Fangyan Ma, Xue Chen. National Cybersecurity Capability Classification and Crime Vulnerability Correlation Analysis: Entropy-Weighted WAM-K-Means Clustering Application. Academic Journal of Computing & Information Science (2025), Vol. 8, Issue 10: 37-43. https://doi.org/10.25236/AJCIS.2025.081006.

[1] Wang Juanjuan,Zhang Qian. A study of major global cybersecurity indices[J]. Information and Communication Technology and Policy,2024,50(08):2-8.

[2] Wang Na, Zhang Xinhai, Chang Yamin. Cyber security posture prediction based on data decomposition and multi-model switching[J/OL]. Computer Science and Exploration,1-14[2025-01-28]. http://kns.cnki.net/kcms/detail/11.5602.tp.20241211.1604.006.html.

[3] ZHAO Di, CHEN Peng, JIANG Huan, et al. Research on joint criminal network and influencing factors based on geographical characteristics of offenders[J]. Geography and Geographic Information Science,2022,38(05):57-64.

[4]] Liu Yunxiao. Research on the investigation of criminal cases involving fraudulent network blackmail [D]. People’s Public Security University of China, 2023.DOI:10.27634/d.cnki.gzrgu.2023.000167.

[5] Shiqi Zhang. A study on the harmonization of the definition of cyber terrorism crime[D]. East China University of Politics and Law,2022.DOI:10.27150/d.cnki.ghdzc.2022.000190.

[6] Xiang Yimeng. The Realistic Dilemma of International Cooperation on Transnational Cybercrime and China’s Response[D]. Zhongnan University of Economics and Law, 2023. DOI:10.27660/d.cnki. gzczu.2023.001587.