Privacy Preserving Data Stream Mining Using Two Phase Geometric Data Perturbation

Data mining is an information technology that extracts valuable knowledge from large amounts of data. Recently, data streams are emerging as a new type of data, which are different from traditional static data. The characteristics of data streams are: Data has timing preference; data distribution changes constantly with time; the amount of data is enormous. Data flows in and out with fast speed; and immediate response is required To preserve data privacy during data mining, the issue of privacy- preserving data mining has been widely studied and many techniques have been proposed. However, existing techniques for privacy-preserving data mining are designed for traditional static databases and are not suitable for data streams. So the privacy preservation issue of data streams mining is a very important issue. This work is about proposing a Method and algorithms for the process of Geometric Data Perturbation or Geometric Data Transformation to achieve privacy preservation. Geometric Data Perturbation is a kind of data perturbation techniques. In this report, we describe the geometric transformations including translation, scaling, rotation, which can transform data in the protection of privacy while maintaining the similarity between data objects.