Comprehensive Research on Privacy Preserving Emphasizing on Distributed Clustering

Often, the information is sensitive or private in nature and these sensitive data when mined violates the privacy of the individuals. Privacy preserving data mining (PPDM) mines the data but intends to preserve the privacy of susceptible data without ever actually seeing it. This paper recaps the important techniques in PPDM like anonymization, perturbation and cryptography. Nowadays, data mining is extensively used when the data is distributed among multiple parties. This paper highlights the research carried out in privacy preserving distributed clustering. Clustering is an effective method to discover data distribution and patterns in datasets. Significant research in privacy preserving distributed clustering is shaped on k-means clustering algorithm with secure multiparty computation (SMC). This work focuses on the previous development, existing challenges, and upcoming trends in privacy preserving k- means clustering with horizontally and vertically distributed data.