DP-WHERE: Differentially private modeling of human mobility

摘要: Models of human mobility have broad applicability in urban planning, ecology, epidemiology, and other fields. Starting with Call Detail Records (CDRs) from a cellular telephone network that gone through straightforward anonymization procedure, the prior WHERE modeling approach produces synthetic CDRs for population. The accuracy has been validated against billions location samples hundreds thousands cell phones New York Los Angeles metropolitan areas. In this paper, we introduce DP-WHERE, which modifies by adding controlled noise to achieve differential privacy, strict definition privacy makes no assumptions about power or background knowledge potential adversary. We also present experiments showing DP-WHERE remains close real CDRs. With work, aim enable creation possible release models capture patterns populations while preserving privacy.