Research on Cybercrime Prevention and Control Strategies Based on K-means++ and PSO Algorithm

This paper focuses on the field of cybercrime, analysing its global pattern and the effectiveness of policy responses through a multimethod approach. The World Cybercrime Index (WCI) is constructed with the help of experts' experience, which reveals the characteristics of the global cybercrime risk distribution, i.e. Europe and North America have a high prevalence of cybercrime, followed by Asia, and South America and Africa have a low prevalence of cybercrime. The K-means++ clustering algorithm is used to classify the risk of 97 countries/regions, and the results match the actual distribution. In the research of policy effectiveness, we constructed a performance score index and found that there is a mutually reinforcing relationship between cybercrime risk and the level of cybersecurity construction; we defined the security index S, and constructed a regression model by combining the political data and legal density of the ITU, and concluded that the legal measures are the most effective in improving the security index. The particle swarm optimisation algorithm is used to explore the optimal political scenarios, provide decision-making reference for legislators, and analyse the reasons for the differences in legal density, and the research results have important reference value for the formulation of global cybercrime prevention and control policies.