Privacy in Statistical Databases

1. Front Matter

   PDF

2. Tabular Data Protection

   1. Using a Mathematical Programming Modeling Language for Optimal CTA

      • Jordi Castro, Daniel Baena

      Pages 1-12

   2. A Data Quality and Data Confidentiality Assessment of Complementary Cell Suppression

      • Lawrence H. Cox

      Pages 13-23

   3. Pre-processing Optimisation Applied to the Classical Integer Programming Model for Statistical Disclosure Control

      • Martin Serpell, Alistair Clark, Jim Smith, Andrea Staggemeier

      Pages 24-36

   4. How to Make the τ-ARGUS Modular Method Applicable to Linked Tables

      • Peter-Paul de Wolf, Sarah Giessing

      Pages 37-49

   5. Bayesian Assessment of Rounding-Based Disclosure Control

      • Jon J. Forster, Roger C. Gill

      Pages 50-63

   6. Cell Bounds in Two-Way Contingency Tables Based on Conditional Frequencies

      • Byran Smucker, Aleksandra B. Slavković

      Pages 64-76

   7. Invariant Post-tabular Protection of Census Frequency Counts

      • Natalie Shlomo, Caroline Young

      Pages 77-89

3. Microdata Protection: Methods and Case Studies

   1. A Practical Approach to Balancing Data Confidentiality and Research Needs: The NHIS Linked Mortality Files

      • Kimberly Lochner, Stephanie Bartee, Gloria Wheatcroft, Christine Cox

      Pages 90-99

   2. From t-Closeness to PRAM and Noise Addition Via Information Theory

      • David Rebollo-Monedero, Jordi Forné, Josep Domingo-Ferrer

      Pages 100-112

   3. Robustification of Microdata Masking Methods and the Comparison with Existing Methods

      • Matthias Templ, Bernhard Meindl

      Pages 113-126

   4. A Preliminary Investigation of the Impact of Gaussian Versus t-Copula for Data Perturbation

      • Mario Trottini, Krish Muralidhar, Rathindra Sarathy

      Pages 127-138

   5. Anonymisation of Panel Enterprise Microdata – Survey of a German Project

      • Maurice Brandt, Rainer Lenz, Martin Rosemann

      Pages 139-151

4. Microdata Protection: Disclosure Risk Assessment

   1. Towards a More Realistic Disclosure Risk Assessment

      • Jordi Nin, Javier Herranz, Vicenç Torra

      Pages 152-165

   2. Assessing Disclosure Risk for Record Linkage

      • Chris Skinner

      Pages 166-176

   3. Robust Statistics Meets SDC: New Disclosure Risk Measures for Continuous Microdata Masking

      • Matthias Templ, Bernhard Meindl

      Pages 177-189

   4. Parallelizing Record Linkage for Disclosure Risk Assessment

      • Joan Guisado-Gámez, Arnau Prat-Pérez, Jordi Nin, Victor Muntés-Mulero, Josep Ll. Larriba-Pey

      Pages 190-202

   5. Extensions of the Re-identification Risk Measures Based on Log-Linear Models

      • Daniela Ichim

      Pages 203-212

   6. Use of Auxiliary Information in Risk Estimation

      • Loredana Di Consiglio, Silvia Polettini

      Pages 213-226

   7. Accounting for Intruder Uncertainty Due to Sampling When Estimating Identification Disclosure Risks in Partially Synthetic Data

      • Jörg Drechsler, Jerome P. Reiter

      Pages 227-238

Privacy in statistical databases is a discipline whose purpose is to provide solutions to the tension between the increasing social, political and economical demand of accurate information, and the legal and ethical obligation to protect the privacy of the various parties involved. Those parties are the respondents (the individuals and enterprises to which the database records refer), the data owners (those organizations spending money in data collection) and the users (the ones querying the database, who would like their queries to stay con?d- tial). Beyond law and ethics, there are also practical reasons for data collecting agencies to invest in respondent privacy: if individual respondents feel their p- vacyguaranteed,they arelikelyto providemoreaccurateresponses. Data owner privacy is primarily motivated by practical considerations: if an enterprise c- lects data at its own expense, it may wish to minimize leakage of those data to other enterprises (even to those with whom joint data exploitation is planned). Finally, user privacy results in increased user satisfaction, even if it may curtail the ability of the database owner to pro?le users. Thereareatleasttwotraditionsinstatisticaldatabaseprivacy,bothofwhich started in the 1970s: one stems from o?cial statistics, where the discipline is also known as statistical disclosure control (SDC), and the other originatesfrom computer science and database technology. In o?cial statistics, the basic c- cern is respondent privacy.

• anonymisation
• data confidentiality
• data mining
• data perturbation
• data quality
• database
• information retrieval
• microdata masking
• microdata protection
• on-line databases
• p2p
• privacy
• tabular data protection

Policies and ethics