Coordination Approach to Mitigate Composition Attack

Data anonymization is one of the primary techniques, which is used in privacy preserving data publishing. Numerous methods have been proposed to anonymize data sets, that are used by different organizations to publish their data sets independently. However, in actuality, a data set is infrequently isolated, and data sets published by multiple organizations may obtain a record of the same person. An individually published data set may have a small privacy risk. However, the combination of two published data sets may breach individuals' privacy, who are common in both data sets by composition attack. A few researches have been carried out to prevent it without considering coordination among multiple organizations before data publishing. Though coordination is an excellent way to avoid such problems, organizations do not want to share their customer data sets with others. In this work, we introduce a coordination method where complete data sharing is not necessary. This will effectively overcome privacy concerns. It has been shown experimentally that the proposed technique anonymizes data set to satisfy the privacy requirement and preserves high data utility.