CAJ | 학술논문

KACA算法是一种较好的基于局部重编码的k-匿名化算法,它产生的匿名化微数据的信息损失少,数据可用性高.但该算法的效率低,不适合匿名化规模很大的微数据.为解决该问题,将高效的聚类算法--c-modes算法--与KACA算法结合,提出了c-modes-KACA算法.该算法首先采用C-modes算法将整个数据集划分为较大的类,然后采用KACA算法来分别k-匿名化这些类中元组数大于2k-1的类.实验表明,c-modes-KACA算法产生的信息损失量与KACA算法相近,效率却比KACA算法高得多.
KACA산법시일충교호적기우국부중편마적k-닉명화산법,타산생적닉명화미수거적신식손실소,수거가용성고.단해산법적효솔저,불괄합닉명화규모흔대적미수거.위해결해문제,장고효적취류산법--c-modes산법--여KACA산법결합,제출료c-modes-KACA산법.해산법수선채용C-modes산법장정개수거집화분위교대적류,연후채용KACA산법래분별k-닉명화저사류중원조수대우2k-1적류.실험표명,c-modes-KACA산법산생적신식손실량여KACA산법상근,효솔각비KACA산법고득다.
KACA is a fine k-anonymization algorithm based on local-recoding,which anonymizes microdata with low information loss,so the anonymized microdata has high data utility.However,the KACA algorithm is inefficient in k-anonymzing large datasets.To solve the problem,an efficient k-anonymization algorithm is proposed,which combines the efficient c-modes algorithm with the KACA algorithm.The proposed algorithm adopts the c-modes algorithm to partition the whole dataset into some large clusters at first.then uses the KACA algorithm to k-anonymize these clusters which have more than 2k-1 tuples respectively.Experimental results show that the proposed algorithm has higher efficiency than KACA with similar information loss.