Privacy-preserving Naïve Bayes classification using trusted third party computation over vertically partitioned distributed progressive sequential data streams

Abstract

Privacy preservation in distributed progressive data stream is an active area of research in the present time. In a typical scenario, multiple parties may be wishing to collaborate to extract interesting global information such as class labels without breaching privacy. This may be particularly useful in applications such as customer retention, medical research etc. In the present work, we aim to develop a global classification model based on the Naïve Bayes classification scheme. The Naïve Bayes classification has been used because of its applicability in case of customer retention application such as car evaluation dataset. For privacy preservation of the data, the concept of trusted third party has been used. We have proposed algorithms and tested car evaluation dataset for vertical partitioned progressive sequential data streams. © Springer-Verlag Berlin Heidelberg 2011.