Abstract

Privacy preserving data publishing(PPDF)is an emerging technology in the datamining field which\nperforms datamining operations in a secured manner to preserve the confidential/sensitive information. Preserving\nprivacy while publishing medical information has become an important challenge in this area due to its high\nconfidentiality. While publishing medical data the PPDF scheme should maximize the data utility at the same time\nshould have a minimum data disclosure risk. This paper concerned with privacy of medical data while publishing the\npatient information for research or analysis purposes. K-anonymity and l-diversity are the most popular techniques used\nfor preserving privacy. These techniques does not consider the semantic relationship between the data values so they are\nprone to similarity attack .In this paper, we present a privacy-preserving data publishing framework for publishing large\ndatasets with the goals of providing different levels of utility to the users based on their access privileges. The proposed\nsystem overcome the similarity attack by applying a privacy preservation approach which uses a key attribute masking\ntechnique and an anonymization process.The results showed that the semantic anonymization increases the privacy level\nwith effective data utility.