Kişisel bilgilerin gizlenmesi veri madenciliği

Data Mining allows large database owners to share and extract useful knowledge that could not be deduced with traditional approaches like statistics. However, these sometimes reveal sensitive knowledge or breach individuals' privacy. The term sanitization is given to the process of changing original database into another one from which we can mine without exposing sensitive knowledge. This process should be guided by little distortion on the database. In this dissertation, we address these issues in a data mining branch called Privacy Preserving Data Mining. In particular, we focus on association rule hiding (ARH) and evaluate the heuristic approaches for this purpose. We also apply these heuristic approaches on a number of publicly available datasets and examine the results.