Anonymizing Temporal Data

Abstract

-Temporal data are time-critical in that the snapshot at each timestamp must be made available to researchers in a timely fashion. However, due to the limited data, each snapshot likely has a skewed distribution on sensitive values, which renders classical anonymization methods not possible. In this work, we propose the "reposition model" to allow a record to be published within a close proximity of original timestamp. We show that reposition over a small proximity of timestamp is sufficient for reducing the skewness of a snapshot, therefore, minimizing the impact on window queries. We formalize the optimal reposition problem and present a linear-time solution. The contribution of this work is that it enables classical methods on temporal data.

Privacy Technology to Support Data Sharing for Comparative Effectiveness Research

“…New emerging applications may involve new data types and there might be no privacy standards to protect them. Such a gap between policy and technology calls for substantial future development of new standards of healthcare data privacy protection for genomic data 99–103 , set-valued data 104 , time series data 105 , text data 106,107 , and image data 108 , which have not been adequately studied in the privacy perspective.…”

Guest editorial: web of things

“…In addition, temporal data are time-critical in that the snapshot at each timestamp must be made available in a timely fashion. However, due to the limited data, each snapshot likely has a skewed distribution on sensitive values, which renders classical anonymization methods not possible (Wang et al 2010).…”

Protecting Privacy in Data Release

“…Some of these (limiting) assumptions can be summarized as follows: (1) each respondent is represented by a single tuple in the microdata table; (2) all data to be released are stored in a single table; (3) once released, data are not further modified; (4) all the data that need to be released are available to the data holder before their release; (5) the same degree of privacy is guaranteed to all data respondents; (6) the released microdata table has a single quasi-identifier, known in advance; and (7) no external knowledge (except for that behind linking attacks counteracted by kanonymity) is available to recipients. Recently, the scientific community has started to extend the pioneering techniques illustrated so far in this chapter removing these assumptions, proposing solutions specifically tailored for supporting, among other scenarios: (1) multiple tuples per respondent (e.g., [101,107]); (2) release of multiple tables (e.g., [86,107]); (3) data republication (e.g., [113]); (4) continuous data release (e.g., [71,109,118]); (5) personalized privacy preferences (e.g., [56,112]); (6) multiple and/or non-predefined quasi-identifiers (e.g., [89,101]); (7) adversarial external knowledge (e.g., [21,76,79]). Figure 2.4 summarizes some notable solution recently proposed to extend the definitions of k-anonymity,`-diversity, and tcloseness removing the above-illustrated assumptions.…”