CMAR: Accurate and Efficient Classification Based on Multiple Class-Association Rules
ICDM '01 Proceedings of the 2001 IEEE International Conference on Data Mining
k-anonymity: a model for protecting privacy
International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems
Disclosure Limitation of Sensitive Rules
KDEX '99 Proceedings of the 1999 Workshop on Knowledge and Data Engineering Exchange
Privacy preserving frequent itemset mining
CRPIT '14 Proceedings of the IEEE international conference on Privacy, security and data mining - Volume 14
IEEE Transactions on Knowledge and Data Engineering
Detecting privacy and ethical sensitivity in data mining results
ACSC '04 Proceedings of the 27th Australasian conference on Computer science - Volume 26
Hiding Sensitive Association Rules with Limited Side Effects
IEEE Transactions on Knowledge and Data Engineering
A Max-Min Approach for Hiding Frequent Itemsets
ICDMW '06 Proceedings of the Sixth IEEE International Conference on Data Mining - Workshops
Hiding Sensitive Associative Classification Rule by Data Reduction
ADMA '07 Proceedings of the 3rd international conference on Advanced Data Mining and Applications
Two new techniques for hiding sensitive itemsets and their empirical evaluation
DaWaK'06 Proceedings of the 8th international conference on Data Warehousing and Knowledge Discovery
Associative classification rules hiding for privacy preservation
International Journal of Intelligent Information and Database Systems
| Hi-index | 0.00 |
When data are to be shared between business partners, there could be some sensitive patterns which should not be disclosed to the other parties. On the other hand, the "quality" of the data must also be preserved. This creates an interesting question: how can we maintain the shared data that are guaranteed to have the quality, and the certain types of sensitive patterns be removed or "hidden"? In this paper, we address such the problem of sensitive classification rule hiding by using data reduction approach, i.e. removing the whole selected tuples in the given dataset. We focus on a specific type of classification rules, i.e. associative classification rules. In our context, a sensitive rule is hidden when its support falls below a minimal support threshold. Meanwhile, the impact on the data quality of the dataset is represented in term of a number of false-dropped rules, and a number of ghost rules. We present a few observations on the data quality with regard to the data reduction processes. From the observations, we can represent the impact by each reduction precisely without any re-applying the classification algorithm. Subsequently, we propose a heuristic algorithm to hide the sensitive rules based on the observations. Experimental results are presented to show the effectiveness and the efficiency of the proposed algorithm.