Geometric Data Perturbation Techniques in Privacy Preserving On Data Stream Mining

Data mining is the information technology that extracts valuable knowledge from large amounts of data. Due to the emergence of data streams as a new type of data, data stream mining has recently become a very important and popular research issue. Privacy preservation issue of data streams mining is very important issue, in this dissertation work, an approach based on Geometric data perturbation has been proposed, which extends the existing process of data streams clustering to achieve privacy preservation. Our objective is to reduce the tradeoff between mining accuracy while minimizing information loss when data undergoing the process of perturbation. Experimental results will show that the method not only can preserve data privacy but also can mine data streams accurately. An effective approach of geometric transformation based data perturbation of data stream for mining has been proposed. It aims privacy of sensitive information before release while obtaining accuracy of data streamclustering with minimum information loss.