摘要： 为使微数据发布在满足K-匿名要求的同时提高匿名数据的精度，提出多维泛化路径的概念及相应的2种K-匿名算法，包括完整Filter K-匿名算法和部分Filter K-匿名算法。将它们与Incognito算法和Datafly算法进行比较，实验结果表明2种算法都能有效降低匿名信息损失，提高匿名数据精度和处理效率。

关键词: K-匿名, 微数据, 隐私保护, 域泛化层次结构

Abstract: Microdata publication need satisfy the basic K-anonymity requirement as well as improve the precision of anonymized data. This paper proposes two related K-anonymity algorithms based on the notion of multi-dimensional generalization path, namely K-anonymity Filter algorithm and K-anonymity partial Filter algorithm. In comparison with classic Datafly algorithm and Incognito algorithm, the two algorithms offer more efficiency for both reducing anonymization cost and improving data precision.

Key words: K-anonymity, microdata, privacy protection, domain generalization hierarchy

中图分类号: 

• TP309