Adaptive Voronoi Masking: A Method to Protect Confidential Discrete Spatial Data

Geomasks assure the protection of individuals in a discrete spatial point data set by aggregating, transferring or altering original points. This study develops an alternative approach, referred to as Adaptive Voronoi Masking (AVM), which is based on the concepts of Adaptive Aerial Elimination (AAE) and Voronoi Masking (VM). It considers the underlying population density by establishing areas of K-anonymity in which Voronoi polygons are created. Contrary to other geomasks, AVM considers the underlying topography and displaces data points to street intersections thus decreasing the risk of false-identification since residences are not endowed with a data point.
The geomasking effects of AVM are examined by various spatial analytical results and are compared with the outputs of AAE, VM, and Donut Masking (DM). VM attains the best efficiency for the mean centres whereas DM does for the median centres. Regarding the Nearest Neighbour Hierarchical Cluster Analysis and Ripley’s K-function, DM demonstrates the strongest performance since its cluster ellipsoids and clustering distance are the most similar to those of the original data. The extend of the original data is preserved the most by VM, while AVM retains the topology of the point pattern. Overall, AVM was ranked as 2nd in terms of data utility (i) and also outperforms all methods regarding the risk of false re-identification (ii) because no data point is moved to a residence. Furthermore, AVM maintains the Spatial K-anonymity (iii) which is also done by AAE and partly by DM. Based on the performance combination of these factors, AVM is an advantageous technique to mask geodata.