基于变长编码遗传算法的K-匿名化

K-anonymization Based on Genetic Algorithm of Variable Length Encoding

针对传统遗传编码存在求解效率低且信息失真大的问题,提出一种基于不定长密歇根编码的遗传算法,采用多种启发式策略进行杂交操作,将基于遗传算法的聚类方法应用到K-匿名化问题中。实验结果表明,该方法可以更好地降低信息失真,从而实现K-匿名化问题。

Aiming at the problems that raditional genetic encoding has low efficiency and large information loss, this paper proposes a michigan-based variable-length coding Genetic Algorithm（GA）, which adopts various heuristic strategies to select genes for crossover operation. It is applied to the problem of K-anonymization. Experimental results show this method can further reduce the information loss and it is a new way to solve the problem of K-anonymization.