Translator Disclaimer
13 January 2012 Performance evaluation of various K- anonymity techniques
Author Affiliations +
Today's advanced scenario where each information is available in one click, data security is the main aspect. Individual information which sometimes needs to be hiding is easily available using some tricks. Medical information, income details are needed to be kept away from adversaries and so, are stored in private tables. Some publicly released information contains zip code, sex, birth date. When this released information is linked with the private table, adversary can detect the whole confidential information of individuals or respondents, i.e. name, medical status. So to protect respondents identity, a new concept k-anonymity is used which means each released record has at least (k-1) other records in the release whose values are distinct over those fields that appear in the external data. K-anonymity can be achieved easily in case of single sensitive attributes i.e. name, salary, medical status, but it is quiet difficult when multiple sensitive attributes are present. Generalization and Suppression are used to achieve k-anonymity. This paper provides a formal introduction of k-anonymity and some techniques used with it l-diversity, t-closeness. This paper covers k-anonymity model and the comparative study of these concepts along with a new proposed concept for multiple sensitive attributes.
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Nidhi Maheshwarkar, Kshitij Pathak, and Vivekanand Chourey "Performance evaluation of various K- anonymity techniques", Proc. SPIE 8350, Fourth International Conference on Machine Vision (ICMV 2011): Computer Vision and Image Analysis; Pattern Recognition and Basic Technologies, 83501Y (13 January 2012);


Back to Top