In the era of big data, the collaborative optimization of data integration and privacy protection has become a core challenge in digital governance. Privacy-Preserving Record Linkage (PPRL) technology offers a secure paradigm for integrating data from multiple sources. However, existing multi-criteria evaluation frameworks, while overcoming the limitations of single-indicator assessment, often fail to fully leverage structured expert knowledge and cannot adapt evaluation results to scenario-specific requirements. To address these issues, this study extends Han et al. (2024) ’s purely objective CRITIC framework by proposing a scenario-driven hybrid evaluation mechanism that integrates the Analytic Hierarchy Process (AHP) with the Criteria Importance Through Intercriteria Correlation (CRITIC) method. The framework combines statistical characteristics of the data with structured expert knowledge to mitigate the subjective bias of AHP and alleviate the scenario insensitivity of pure CRITIC, and introduces an expert participation coefficient to dynamically balance the relative contributions of subjective expert preferences and objective data characteristics across different application scenarios. It constructs a scalable evaluation system encompassing multiple indicators, such as runtime, linkage quality, and security, with its core strength lying in the dynamic adjustment of indicator weights based on scenario preferences. Experimental results demonstrate that the framework generates evaluation outcomes that are more aligned with scenario-specific needs and user preferences, compared with purely objective approaches. Overall, this study extends existing purely objective evaluation paradigms by integrating structured expert knowledge. This approach enables the effective identification of PPRL methods that are better suited to real-world scenarios and provides interpretable decision support for the selection and optimization of PPRL technologies in digital governance scenarios.
Han et al. (Tue,) studied this question.