A Multi-Indicator Comprehensive Evaluation and Clustering Analysis of Global Cybercrime Based on Entropy Weight Method and TOPSIS Model

Abstract

This paper constructs a comprehensive evaluation and analysis framework for multi-source indicators to characterize the differences in cybercrime and governance capabilities among countries worldwide. In terms of indicator construction, the study first standardizes the raw variables based on multidimensional evaluation data, introduces the information entropy method to determine indicator weights, and constructs a weighted comprehensive evaluation system to form an overall index. Building on this foundation, the TOPSIS method is employed to rank countries relative to one another and calculate their proximity, thereby enabling comprehensive comparative analysis under multi-indicator conditions. Furthermore, by combining K-means clustering and principal component analysis, the country samples are structurally clustered and visualized in a low-dimensional format to reveal the intrinsic relationships between different levels of governance and technical capabilities. At the same time, correlation analysis is used to characterize the statistical relationships among key variables, thereby enhancing the model’s explanatory power. The research methodology possesses strong scalability and versatility, providing a reference for the quantitative evaluation and comparative analysis of complex, multi-indicator systems.