Research on preservation of proximity privacy of numerical sensitive data
数值型敏感属性的近邻泄露保护方法研究

Proximity breach occurs when an adversary may have high confidence to infer the victim’s value fall in a short interval in publishing numerical sensitive data.In view of such proximity breach in data publishing when sensitive values were numerical,this paper proposed a model called(εp,l)-anonymity to prevent proximity breach on the idea of lossy join.To control similarity of sensitive values in different ranges,this model set different threshold values εi(1≤i≤p).The results of experiments suggest that the model is able to reduce proximity breach apparently and enforces security of data publishing with a better utility.