Thesis
French
ID: <
10670/1.15hwu4>
Abstract
Personal data anonymization requires complex algorithms aiming at avoiding disclosure risk without losing data utility. In this thesis, we describe a model-driven approach guiding the data owner during the anonymization process. The guidance may be informative or suggestive. It helps the data owner in choosing the most relevant algorithm given the data characteristics and the future usage of anonymized data. The guidance process also helps in defining the best input values for the algorithms. In this thesis, we focus on generalization algorithms for micro-data. The knowledge about anonymization is composed of both theoretical aspects and experimental results. It is managed thanks to an ontology.