Anonymization of Multiple and Personalized Sensitive Attributes.

In the past, many algorithms have presented to hide the sensitive information but most of them identify the sensitive information as the same for all users/transactions, which is not a situation happened in realistic applications. In this paper, we present the (k, p)-anonymity framework to hide not only the multiple sensitive information but also the personal sensitive ones. Extensive experiments indicated that the proposed algorithm outperforms the-state-of-the-art algorithms in terms of information loss and runtime.