Original Article

Acta Informatica

, Volume 48, Issue 1, pp 51-66

First online:

Efficient systematic clustering method for k-anonymization

  • Md. Enamul KabirAffiliated withDepartment of Mathematics and Computing, University of Southern Queensland
  • , Hua WangAffiliated withDepartment of Mathematics and Computing, University of Southern Queensland
  • , Elisa BertinoAffiliated withDepartment of Computer Science and CERIAS, Purdue University Email author 

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access


This paper presents a clustering (Clustering partitions record into clusters such that records within a cluster are similar to each other, while records in different clusters are most distinct from one another.) based k-anonymization technique to minimize the information loss while at the same time assuring data quality. Privacy preservation of individuals has drawn considerable interests in data mining research. The k-anonymity model proposed by Samarati and Sweeney is a practical approach for data privacy preservation and has been studied extensively for the last few years. Anonymization methods via generalization or suppression are able to protect private information, but lose valued information. The challenge is how to minimize the information loss during the anonymization process. We refer to the challenge as a systematic clustering problem for k-anonymization which is analysed in this paper. The proposed technique adopts group-similar data together and then anonymizes each group individually. The structure of systematic clustering problem is defined and investigated through paradigm and properties. An algorithm of the proposed problem is developed and shown that the time complexity is in \({O(\frac{n^{2}}{k})}\), where n is the total number of records containing individuals concerning their privacy. Experimental results show that our method attains a reasonable dominance with respect to both information loss and execution time. Finally the algorithm illustrates the usability for incremental datasets.