A Quantitative Model for Evaluating the Effectiveness of Data Desensitization Based on Information Entropy

To assist enterprises in selecting appropriate data desensitization algorithms under specific scenarios, ensuring both data security and usability, this study proposes a quantitative model for evaluating the effectiveness of desensitization algorithms based on information entropy. The model, after analyzing relevant domestic and international research, establishes a comprehensive evaluation system that includes two key dimensions: security and usability. It employs information entropy and loss entropy to quantify the uncertainty and information loss of desensitization algorithms. The model also uses the entropy weight method and the Analytic Hierarchy Process (AHP) to determine the weights of the evaluation indicators, ensuring the objectivity and adaptability of the assessment. Experimental validation has demonstrated the model's effectiveness and rationality, providing enterprises with a scientific tool for the assessment of desensitization technology.