A REVIEW ON ANONYMIZATION TECHNIQUES FOR PRIVACY PROTECTION IN DATA MINING
U. Saranya, A. Logeswari, U. Sujatha · International Journal of Engineering Applied Sciences and Technology · 2020
Data anonymization is a primary technique used purposely for the protection of privacy.To remain anonymous, it excludes personal identifiable information from data sets.Preservation of privacy is a major factor to be considered in protecting against attacks by unauthorized entities on identity disclosure and ensuring that data is anonymized and still efficient for analytical tasks.The data can be released to ensure that the loss of information is minimal in order to maintain the utility of the analysis for further tasks.The data in the dataset may be a mixture of sensitive and non-sensitive information.To order to protect sensitive data from the outside world, privacy protection strategies for data publishing are rapidly increased.K-anonymity is one of important privacy protection strategies, but more attention needs to be paid to increasing data usefulness and more loss of knowledge before publishing.Using L-diversity and Tcloseness strategies, privacy security can be based on nodes.Edge perturbation and edge randomization are privacy conservation techniques in social network graphs. Relational information and social network data protect privacy using K-anonymity, edge perturbation, edge randomization, and L-diversity techniques. In this paper, a comparative review and study on K-Anonymity, L-Diversity, T-Closeness and perturbation Anonymization techniques is presented along with slicing for highdimensional databases and procedure for following the reduction of dimensionality with selection of features.