Published Paper Details:

Sharing of big personal data is usually possible and allowed in various data sharing agreements between organizations/ institutions because the data is anonymized, without containing any identifiable data. Anonymization is done to enforce privacy and to maintain confidentiality. Anonymized data however is much efficient for aggregate statistics and analytics, what if we want to be able to analyze the data at an individual level but maintain the privacy of the individual? In this paper we aim to achieve sharing of both anonymized and identifiable data while maintaining privacy. We do this using the K-Anonymity algorithm and a custom mapping of the anonymized data to the identifiable data. We consider extracting identifiable data into a separate table out of the compatible destination database. Extracting all the quasi-identifiable data and combine them. Then store the concatenated quasi-identifiers in the same separate table as the identifiers.