Unique in the shopping mall: On the reidentifiability of credit card metadata

Abstract

Large-scale data sets of human behavior have the potential to fundamentally transform the way we fight diseases, design cities, or perform research. Metadata, however, contain sensitive information. Understanding the privacy of these data sets is key to their broad use and, ultimately, their impact. We study 3 months of credit card records for 1.1 million people and show that four spatiotemporal points are enough to uniquely reidentify 90% of individuals.We show that knowing the price of a transaction increases the risk of reidentification by 22%, on average. Finally, we show that even data sets that provide coarse information at any or all of the dimensions provide little anonymity and that women are more reidentifiable than men in credit card metadata.Large-scale data sets of human behavior have the potential to fundamentally transform the way we fight diseases, design cities, or perform research. Ubiquitous technologies create personal metadata on a very large scale. Our smartphones, browsers, cars, or credit cards generate infor-

Privacy Preservation in the Age of Big Data: A Survey

“…Later Sweeney showed that 87% (216 million of 248 million) of the population in the United States could be uniquely identified if their 5-digit zip code, date of birth and gender is known [2]. Other work shows that, in general, subjects can be easily and uniquely re-identified using a very sparse subset of their data trails as recorded in commercial databases [7], especially if the data includes location and financial details. The result is that data subjects (the people the data represents) can be re-identified based on data that has had explicit identifiers removed.…”

“…This is especially true as cheap, high-powered computing proliferates. The Census Bureau may need to rethink its privacy 7 The statement of this policy and links to archived PIAs can be found at: (http://www.census.gov/about/policies/privacy/ pia.html). The PIAs also serve to record information-sharing partners (usually other federal agencies) and consent collection practices.…”

“…It also implements privacy impact assessments per the FISMA requirement mentioned above to ensure that any collected PII is both necessary and permitted. 7 They also do some research and publish reports on internal statistical disclosure control practices [45]. These are mainly syntactic manipulation methods.…”

Privacy Threats in E-Shopping (Position Paper)

“…A relevant example of threat 2.3 in Table 1, which may also imply threats 2.1 and 2.2, appears in [8]. From a large set of simply anonymized financial data (without names, addresses or obvious identifiers), [8] shows that it is possible to de-anonymize 90% of the individuals, if the data contain three items: price, when, and where.…”

“…From a large set of simply anonymized financial data (without names, addresses or obvious identifiers), [8] shows that it is possible to de-anonymize 90% of the individuals, if the data contain three items: price, when, and where. Note that it is very likely that the financial data collected by FN contain the three pieces of information.…”