Privacy Preserving Clustering by Hybrid Data Transformation Approach

Provided by: International Journal of Emerging Technology and Advanced Engineering (IJETAE)
Topic: Data Management
Format: PDF
Numerous organizations collect and share large amounts of data due to the proliferation of information technologies and internet. The information extracted from these databases through data mining process may reveal private information of individuals. Privacy preserving data mining is a new research area, which allows sharing of privacy-sensitive data for analysis purpose. In this paper a hybrid data transformation method is proposed for privacy preserving clustering in centralized database environment by taking the advantage of two existing techniques Principle Component Analysis (PCA) and Non Negative Matrix Factorization (NMF). The experimental results proved that the proposed hybrid method protects private data of individuals and also providing valid clustering results.

Find By Topic