Hiding Association Rules by Using Confidence and Support
IHW '01 Proceedings of the 4th International Workshop on Information Hiding
IEEE Transactions on Knowledge and Data Engineering
Hiding Sensitive Association Rules with Limited Side Effects
IEEE Transactions on Knowledge and Data Engineering
| Hi-index | 0.01 |
A new approach for measuring the side effects of a database is presented here. Different data mining techniques are used in various applications. However, the misuse of these techniques may result in the disclosure of sensitive information. A database may contain sensitive information. This sensitive information must be protected. There is always a risk in releasing a database to the outer world seeking information. The research in advanced data mining techniques also increased the possibility of mining the hidden sensitive information. In this paper, an approach that modifies the transactions in database to decrease confidence of the sensitive rules chosen and measuring the side effects generated during rule hiding, is presented. This approach hides the sensitive rules from the database. The rule hiding process hides all the user chosen sensitive rules. While hiding the sensitive rules, there are obvious side effects, like some of rules may be lost, some of the rules may be newly generated. This approach measures the count of all the constraints for different minimum support threshold (MST) and minimum confidence threshold (MCT) value sets.