Reconstruction Methods for Providing Privacy in Data Mining

Abstract

Data mining is the process of finding correlations or patterns among the dozens of fields in large database. A fruitful direction for data mining research will be the development of techniques that incorporate privacy concerns. Since primary task in our paper is that accurate data which we retrieve should be somewhat changed while providing to users. For this reason, recently much research effort has been devoted for addressing the problem of providing security in data mining. We consider the concrete case of building a decision tree classifier from data in which the values of individual records have been reconstructed. The resulting data records look very different from the original records and the distribution of data values is also very different from the original distribution. By using these reconstructed distribution we are able to build classifiers whose accuracy is comparable to the accuracy of classifiers built with the original data.