APPLICATION OF SIMULATED COMPUTER SIMULATION TO THE TASK OF PERSONAL DEPERSONALIZATION DATA. MODEL AND ALGORITHM FOR DECONTAMINATION BY SYNTHESIS

The second part of the study on the topic of automated depersonalization of personal data is presented. The review and analysis of the prospects for research, performed earlier, is supplemented here by a practical result. A model of the depersonalization process is proposed, reducing task of ensuring anonymity of personal data to manipulation of samples of different types of random elements. Accordingly, the key idea of transforming data to ensure their anonymity, provided that utility is maintained, is to apply the synthesis method, i.e. complete replacement of all unpublished data with synthetic values. The proposed model identifies a set of element types for which synthesis patterns are proposed. The set of patterns compiles the depersonalization algorithm by the synthesis method. Methodically, each template is based on a typical statistical tool – frequency probability estimates, nuclear Rosenblatt-Parsen density estimates, statistical averages and covariances. The application of the algorithm is illustrated by a simple example from the field of civil air transportation.