Privacy-preserving in collaborative data publishing provides methods and tools for publishing the data while protecting the sensitive information in the data set. The success of data mining in privacy relies on the information sharing and quality of data in a distributed environment. Several anonymization techniques have been proposed such as bucketization, generalization which does not prevent...

Preserving privacy while publishing social network data has become a serious issue with the rapid growth of Social Networks. In this work, we propose a perturbation based approach for privacy preserving publication of social network graphs and evaluate the utility aspect of our proposed method using real world dataset.

Many databases contain data about individuals that are valuable for research, marketing, and decision making. Sharing or publishing data about individuals is however prone to privacy attacks, breaches, and disclosures. The concern here is about individuals’ privacy—keeping the sensitive information about individuals private to them. Data mining in this setting has been shown to be a powerful to...

Current privacy preserving methods in data publishing always remove the individually identifying attribute first and then generalize the quasi-identifier attributes. They cannot take the individually identifying attribute into account. In fact, tuples will become vulnerable in the situation of multiple tuples per individual. In this paper, we analyze the individually identifying attribute in th...