Privacy Preserving for Micro Data using Overlap Slicing

In recent years several anonymization techniques are proposed for preserving the privacy of micro data. Micro data holds information about an individual along with sensitive information. Existing anonymization techniques are bucketization, generalization and slicing. Generalization loses information and does not support high dimensional data. Bucketization does not preserve identity disclosure. The slicing method provides more privacy and data utility than other existing methods but still it has few limitations. To overcome the limitations a new technique called overlap slicing is proposed. Overlap slicing provides a high level of privacy and better data utility than slicing technique. In this method attributes are duplicated in more than one column to confuse opponents and preserve membership disclosure.